CA3159958A1 - System and method for utilizing a portable phototherapy device for the treatment of psoriasis - Google Patents

Abstract

Abstract A system and method for utilizing a portable phototherapy device for the treatment of psoriasis. A portable narrow-band ultraviolet-B (UVB) device is used to detect and treat different skin diseases such as psoriasis. This device has the combination of both hardware and software. It includes a sharp distance sensor, a camera for image processing, a Bluetootht module for wireless connectivity and an IoT program. The camera takes pictures of a person's skin, then the severity of the disease is determined through it. This device includes artificial intelligence to diagnose psoriasis and intelligent control for dosage rate. This device also has image processing to show the improvement of the skin. It should be noted that ultraviolet B (UVB) radiation increases cutaneous vitamin D synthesis. Therefore, treatment with the above device leads to an increase in vitamin D absorption which can help boost the immune system to fight other diseases and viruses. Date Recue/Date Received 2022-05-20 A system and method for utilizing a portable phototherapy device for the treatment of psoriasis. A portable narrow-band ultraviolet-B (UVB) device is used to detect and treat different skin diseases such as psoriasis. This device has the combination of both hardware and software. It includes a sharp distance sensor, a camera for image processing, a Bluetooth module for wireless connectivity and an IoT program. The camera takes pictures of a person's skin, then the severity of the disease is determined through it. This device includes artificial intelligence to diagnose psoriasis and intelligent control for dosage rate. This device also has image processing to show the improvement of the skin. It should be noted that ultraviolet B (UVB) radiation increases cutaneous vitamin D synthesis. Therefore, treatment with the above device leads to an increase in vitamin D absorption which can help boost the immune system to fight other diseases and viruses.

Description

SYSTEM AND METHOD FOR UTILIZING A PORTABLE PHOTOTHERAPY DEVICE FOR
THE TREATMENT OF PSORIASIS
Cross Reference to Related Applications [0001] The application claims priority to and the benefit of US Provisional Patent Application Serial No.
63/191056, entitled "SYSTEM AND METHOD FOR UTILIZING A PORTABLE PHOTOTHERAPY
DEVICE FOR THE TREATMENT OF PSORIASIS" filed on May 20, 2021.
Background

[0002] The embodiments described herein relate to phototherapy technology for treating psoriasis.

[0003] Psoriasis is one of the most common human skin diseases and appears to have an important genetic basis. It is a chronic, healing, recurrent scaling skin disease that affects 1 to 3 percent of the world's population. The diagnosis is made on a clinical basis, although histological examination of the skin biopsy specimen may be helpful. Psoriasis is a debilitating, though rarely life-threatening, disease with social and economic implications that is underappreciated by physicians and other health care providers. Recently, there has been progress in understanding the pathogenesis of psoriasis, and therapeutic advances have improved the care of even the most severely injured patients.

[0004] Psoriasis is a common, chronic, non-communicable skin disease with no clear cause or cure, causing red, itchy, scaly patches. Around 125 million individuals are affected worldwide, making psoriasis a serious global problem with a reported prevalence ranging between 0.09% and 11.43%. According to the WHO Global report on psoriasis, many people suffer from this skin disease due to delayed diagnosis, inadequate treatment, and insufficient access to care. There is an upward trend in the prevalence of psoriasis across countries. Data from the National Health and Nutrition Examination Survey indicated an increase in prevalence from 1.62% in 2004 to 3.10% by 2010 in the United States. The rising prevalence of psoriasis drives the need for psoriasis treatments.

[0005] Fortunately, psoriasis is characterized by overgrowth and abnormal keratinocyte differentiation, but it is completely reversible with proper treatment. The trigger for keratinocyte responses is thought to be activation of the cellular immune system; T cells, dendritic cells, and immune-related cytokines and chemokines are involved in pathogenesis.

[0006] Psoriasis may get worse with age. The degree of severity depends on hereditary and environmental Date Recue/Date Received 2022-05-20 factors. Its severity ranges from a few red plaques and scattered scales to involvement of almost the entire surface of the body. Recent data shows that the incidence of psoriasis varies according to geographical area;
it is more common in countries far from the equator. Studies on the prevalence and incidence of psoriasis have helped to better understand the burden of the disease.

[0007] Psoriasis can be treated by using various topical and biologic drug treatments. But each has its own properties or limitations. For example, pills, ointments, salicylic acid, bandages, etc., can stop the disease but not cure it. There are other options like oral or injected drugs, skin grafting or genetic modification, and everything related to it, which is used for the treatment of skin diseases.

[0008] There are different types of treatments used for controlling the symptoms, such as topical and systemic therapies as well as phototherapy. Each method has its pros and cons.
However, topical and systemic therapies are known to have a wide array of unintended side effects and immunosuppression.
Furthermore, the global market for psoriasis drugs is declining at a CAGR of -10.8% due to COVID-19's lockdown measures and restrictions of producing non-emergency medicine. Such a trend drives more demand toward alternative treatment methods such as hand-held phototherapy devices that can be administered at home and produced without many medical resources. As such, the preferred method of treatment nowadays is phototherapy with the use of narrowband Ultraviolet B
(NB-UVB) radiation rays.

[0009] It can be practiced in clinics with specialized phototherapy machines, requiring two to three applications per week. However, busy schedules make it hard for patients to balance their 9-5 workdays with their mandatory phototherapy sessions occurring multiple times a week at a clinic. Moreover, due to COVID-19 pandemic, patients are prohibited from in-person doctor visits for non-urgent diseases in order to limit the risk of developing or spreading the virus. Such instances lead to the development of a home-based phototherapy device that users can apply at their convenience. However, the self-application of UVB
rays is susceptible to human errors. Improper and infrequent use can increase the risk of skin cancer, skin thinning, skin burns, and treatment failure.

[0010] Light therapy is a first-line treatment for moderate to severe psoriasis, either alone or in combination with medication. It involves exposing the skin to a controlled amount of natural or artificial light. Repeated treatments are necessary. Therefore, the patient can talk to his / her doctor about whether home phototherapy is an option for him / her.

[0011] There is a desire to provide a safe, reliable, and cost-effective home-based phototherapy device to help patients treat psoriasis. There is a desire for a portable phototherapy device, using Ultraviolet B (UVB) radiation, for the treatment of psoriasis. There is a further desire to use artificial intelligence to assist in the Date Recue/Date Received 2022-05-20 diagnosis of different types of psoriasis and intelligently control the dosage rate.
Summary

[0012] The present innovation is a newer portable type of narrow-band ultraviolet-B (UVB) device, especially a type of targeted phototherapy in terms of medical and cosmetic treatment. The UVB device is an apparatus with a new approach to detecting and treating some skin diseases such as different types of psoriasis (plaque, Pustular, flexural, Guttate, and Erythroderma), Vitiligo, Pityriasis Rosea, Atopic eczema, Hyper bilirubin, and other diseases on different skin types. One of the advantages and technological parts of this device is image processing. When a person is treated, his skin gradually improves, and the image processing shows this improvement, so a lower dose is needed. All devices developed for the treatment of psoriasis have the same function and do not differ from each other.

[0013] This device includes artificial intelligence to diagnose five types of psoriasis and intelligent control for dosage rate. This device also has image processing that can be helpful to show the improvement of the skin. It should be noted that ultraviolet B (UVB) radiation increases cutaneous vitamin D synthesis.
Therefore, treatment with the above device leads to an increase in vitamin D
absorption which can help boost the immune system to fight other diseases and viruses (e.g., coronavirus).

[0014] Furthermore, this device has a combination of both hardware and software. It includes a sharp distance sensor, a camera for image processing, a Bluetooth0 module for wireless connectivity and an IoT
program. The camera takes pictures of a person's skin, then the severity of the disease is determined through it.
Brief Description of the Drawings

[0015] FIG. lA is an exploded view of an assembly of a portable UVB device.

[0016] FIG. 1B is a table listing different component of a portable UVB
device.

[0017] FIG. 1C is a back perspective view of a portable UVB device.

[0018] FIG. 1D is a front perspective view of a portable UVB device.

[0019] FIG. 2A and 2B are diagrams illustrating operation of a portable UVB
device.

[0020] FIG. 3 is a diagram illustrating an exemplary graphical user interface (GUI) for RULA analysis software.

Date Recue/Date Received 2022-05-20

[0021] FIG. 4 is a diagram illustrating an exemplary circuit board of a portable UVB device.

[0022] FIG. 5 diagram illustrating a pre-trained MobileNet Deep CNN
classification system.

[0023] FIG. 6 diagram illustrating 86% Accuracy in Classifying Psoriasis Using the MobileNet Deep Neural Network and Augmentation.

[0024] FIG. 7 is a diagram illustrating a Confusion Matrix.

[0025] FIG. 8A is a table illustrating various evaluation criteria for model testing.

[0026] FIG. 8B is a table illustrating two related outputs of each sample class.

[0027] FIG. 8C is a table illustrating appropriate dose for each skin color to treat psoriasis.

[0028] FIG. 8D is a table illustrating appropriate dose to use in corona.

[0029] FIG. 9 is a diagram illustrating Client Application overview.

[0030] FIG. 10 is a diagram illustrating Doctor Application overview.
Detailed Description

[0031] Psoriasis is a chronic and noncontagious skin disease, which produces red, itchy, and raised patches on the skin. Psoriasis can be treated by applying various treatments including topical therapies, biologic drug treatments, systemic therapies and hospital-based phototherapy. Each has its own characteristics or limitation.
Psoriasis Treatment Options:

[0032] Topical therapies, including creams and ointments applied on top of the affected skin, are popular alternatives for mild cases as they are effective in treating symptoms.
However, they are not a substitute for the treatment that healthcare providers recommend. One major drawback regarding topical treatments is that many contain chemicals, fragrances, and dyes that can irritate the skin and create discomfort or a worsening of symptoms. Additional side effects can include an increased risk for skin cancer, lymphoma, and increased sensitivity to light.

Date Recue/Date Received 2022-05-20

[0033] Systemic therapies include the use of biologics and are taken orally or through injections. This includes drugs such as Remicade0, Humira0, and Enbre10, among others. Systemic therapies are also not recommended for patients who find themselves immunocompromised, pregnant, or breastfeeding.

[0034] The use of biologic drug treatments and injections, is only prescribed for brief periods of time, often alternating with other treatments due to their severe side effect, namely osteoporosis, cancer, digestive and kidney issues. Another factor to note about biologics is that patients are seeing an increase in out-of-pocket costs associated with the pharmaceutical purchases of drugs as some insurance companies have revoked their deductible status, costing patients thousands of dollars.

[0035] Receiving phototherapy treatment at a clinic is a safe yet costly and inconvenient alternative.
Including travel expenses and lost wages, three months of in-office UVB
phototherapy can cost over $1000, making it more expensive per treatment than home-based phototherapy.
Furthermore, clinical-based treatments have waiting times associated with seeing medical specialists in certain countries.
Phototherapy as a Recommended Treatment:

[0036] From all psoriasis treatments available, phototherapy remains the safest option, according to the National Psoriasis Foundation, due to having fewer side effects than biologics. According to the National Psoriasis Foundation, biologics have a chance of increasing infections, which are then required to be monitored to prevent any adverse effects from occurring. Not only is phototherapy safe, but it is also considered to be as effective or more effective than other treatments, according to a published article in the Journal of the American Academy of Dermatology.

[0037] The main differentiators when it comes to home-based and hospital-based phototherapy are cost and accessibility. Home-based units are a one-time purchase that can save patients time, money, and the hassle of frequent clinic visits. Home-based solutions allow patients to manage their schedules without stress as their weekly treatments can be conducted in the comfort of their own home at their self-scheduled times, avoiding conflict with other commitments such as work.

[0038] Patients also avoid the extremely long wait times associated with receiving treatment in Canada, enabling them to consistently treat their condition. In 2013, an economic analysis also concluded that purchasing a home-based phototherapy solution was more cost-effective for patients living 20 miles or more from a clinic, regardless of whether the device was covered by insurance.
Date Recue/Date Received 2022-05-20 Ultraviolet B (UVB) Phototherapy Device:

[0039] An Ultraviolet B (UVB) phototherapy device is an apparatus with a new approach for detecting and treatment of skin disorders, such as Psoriasis (Plaque, Pustular, Flexural, Guttate, and Erythrodermic), Vitiligo, Pityriasis rosea, Atopic eczema, Hyper bilirubin and other diseases on all skin types.

[0040] Ultraviolet (UV) Phototherapy is a method to prevent and cure diseases using artificial ultraviolet radiation. Devices with ultraviolet light function have been developed to treat psoriasis. However, due to their special functionality, these devices require a skilled operator.
Furthermore, they are difficult to use.
Existing devices developed to treat psoriasis have the same function and are not different from each other.

[0041] The present innovation is a newer portable type of narrow-band ultraviolet-B (UVB) device, especially a type of targeted phototherapy in terms of medical and cosmetic treatment. The UVB device is an apparatus with a new approach to detecting and treating some skin diseases, such as psoriasis (plaque, Pustular, flexural, Guttate, and Erythroderma), Vitiligo, Pityriasis Rosea, Atopic eczema, Hyper bilirubin, and other diseases on all skin types.

[0042] One of the advantages and technological parts of this device is image processing. When a person is treated, his skin gradually improves and the image processing shows this improvement, so a lower dose is needed. All devices developed for the treatment of psoriasis have the same function and do not differ from each other.

[0043] The present innovation is a newer portable type of ultraviolet B (UVB) device called narrowband UVB (NB-UVB), especially a kind of targeted phototherapy, one which is in contact with medical and beauty treatment. Therefore, the targeted customers are everyone who has a skin disease, including individuals, beauty clinics and hospitals.

[0044] The narrowband UVB produces the part of the ultraviolet light spectrum band that can be helpful for psoriasis. The lamp of UVB psoriasis using that feature can affect the skin on the cellular level. UVB
light is radiated in a way that can solve the mentioned problem. The UV
radiation causes the cells to mutate and then incomplete metabolism. Therefore, this disease can be treated by killing cells.

[0045] The device has a combination of both hardware and software. This includes a sharp distance sensor, a camera for image processing, and IoT program. To use it, it can be easily placed against the affected area.
When the device is turned on, the user is asked to select their skin type or its color. It also asks which part of body he / she wants to treat. The device has the ability to ask the user how many sessions of treatment it Date Recue/Date Received 2022-05-20 is. On the other hand, it has a Bluetooth0 module and a camera. The camera takes pictures of a person's skin, then determines the severity of the disease through image processing. In the last step, all the information is entered into the table in order to make a decision about a suitable dose of UVB ray that should radiate on a person's skin. The distance of the skin from the device is constantly being measured. If a person moves his/her hand, the sharp distance sensor constantly senses the distance.

[0046] The mentioned device can be used for treating all parts of the body, such as the scalp. The scalp is one of the most difficult areas to treat because the hair hinders the direct effect of ultraviolet irradiation. It has a special comb attachment that can also be used as a distance guard to treat other parts. Consequently, in the device, four parameters are examined: type of disease, skin type, severity of the disease and treatment session. The obtained results are in accordance with the parameters that dermatologists use as a standard criterion for the diagnosis of psoriasis. It can be noted that this portable device has 2 combs on each side of it for putting aside hair.

[0047] This device includes artificial intelligence (Al) to diagnose up to five types of psoriasis and intelligent control for dosage rate. This device also has image processing that can be helpful to show the improvement of the skin.

[0048] Additionally, there is a touched LCD for showing the results and whatever is related. In fact, one of the advantages and technological parts of this device is image processing.
When a person is treated, his skin gradually improves and the image processing shows this improvement, so a lower dose is needed.

[0049] It should be noted that ultraviolet B (UVB) radiation increases cutaneous vitamin D synthesis.
Therefore, treatment with the above device leads to an increase in vitamin D
absorption. Obviously, vitamin D can be helpful in boosting your immune system to fight coronavirus.
Device Hardware:

[0050] The device is a hand-held device equipped with a UVB lamp, a sharp-distance sensor, a camera for image processing and an LCD touch screen. The device emits UVB light that can be used to treat individual parts of the user's body, reducing inflammation and slowing the production of skin cells. This helps manage and prevent a psoriasis flare. The sharp-distance sensor measures the distance of the skin from the device, allowing it to automatically adjust the dosage and exposure time based on your motions. This feature reduces the risk of skin burns, freckles formation, and increased sun sensitivity associated with overexposure to UVB narrowband. The camera will take pictures of the affected skin, determining four parameters: type of disease, skin type, the severity of the disease, and treatment session via image Date Recue/Date Received 2022-05-20 processing. Once the device completes the diagnosis, it will show results and relevant user instructions on the LCD touch screen. The device also comes with two comb attachments used for treating the scalp area and protecting unaffected skin from exposure to UVB rays.

[0051] FIGS. lA to 1D are diagrams illustrating different parts of a portable UVB device assembly including a top cover, LCD, UV protector, ballast base, UV lamp, ballast, power switch, fan, board, lower cover, glass talc, ac socket, comb, camera, distance sensor, and the base of the distance sensor. According to these figures, FIG. lA is an exploded view of an assembly of a portable UVB
device. FIG. 1B is a table listing different components of a portable UVB device. FIG. 1C is a back perspective view of a portable UVB device. FIG. 1D is a front perspective view of a portable UVB device.

[0052] This device has the maximum dimensions of 220 mm (length) X 110 mm (width) X 70 mm (depth) and weighs 0.8 0.1 kg, which allows the user to easily carry and operate it with one hand. According to FIGS. lA and 1B, further details of these components are as follows.

[0053] TOP COVER: This frame is used to protect the components inside the device, which is made of plastic and its minimum and maximum operating temperatures range from minus 10 to 300 degrees Celsius.
The device is sensitive to chloroform, dichloromethane, acetone, benzene, etc.
It will be damaged if used.

[0054] LCD: It is a 3.2-inch color screen that displays device status and treatment progress, and is equipped with a resistive touch screen. The screen resolution is 320 x 240 pixels and 64 thousand colors.
The following details show the characteristics of the LCD:
= 3.2-inch color screen, support 16BIT RGB 65K color display, displays rich colors.
= 240x320 resolution for clear display.
= Support 8-bit/16-bit parallel bus switching (default is 16-bit), fast transmission speed.
= Supports ALIENTEK 5TM32 Mini, Elite, Warship, Explorer, and Apollo development boards for direct plug-in use.
= Supports touch function.
= Supports SD card function expansion.
= Provides a comprehensive sample program for 5TM32 and C51 platforms.
= Military process standards, long-term stability.
= Provides underlying technical support for drivers.

[0055] UV PROTECTOR: The UV Protector is designed to protect the lamp and prevent light radiation to the internal components, as well as to use the maximum power of the lamp for the treatment.

[0056] BALLAST BASE: It is used to hold the ballast in the unit.

Date Recue/Date Received 2022-05-20

[0057] UV LAMP: This lamp is type c and is used for treatment. The UV lamp installed in device is PL -S 9W/01/2P manufactured by Philips . The UV lamp has biological safety - IEC
62471 - Risk Group 3.
The following details show the characteristics of the UVB lamp:
1. General Information = Cap-Base: G23 [ G231 = Life to 50% Failures (Nom): 1000 h = Useful Life (Nom): 1000 h 2. Light Technical = Color Code: 01 = Color Designation: Ultra Violet B
= UV Depreciation at 500 h: 15%
= UV Depreciation at 1000 h: 20%
3. Operating Electrical = Power (Nom): 8.6W
= Lamp Current (Nom): 0.17 A
= Voltage (Nom): 60 V
4. Mechanical and Housing = Cap-Base Information: 2 Pins 5. Approval and Application = Mercury (Hg) Content (Nom): 3.0 mg 6. UV
= UV-B Radiation 100 hr. (IEC): 1.0 w = UV-B Radiation 5hr (IEC): 1.2W

[0058] BALLAST: In a fluorescent lighting system, the ballast controls the current to the lamps and provides sufficient voltage to start the lamps. Without a ballast to limit the current, a fluorescent lamp connected directly to a high voltage source would increase its current draw rapidly and uncontrollably.
Within a second, the lamp would overheat and burn out. During lamp start-up, the ballast must supply high voltage for a short time to create an arc between the two lamp electrodes.
Once the arc is established, the ballast quickly reduces the voltage and regulates the electrical current to produce a uniform light output.

Date Recue/Date Received 2022-05-20 The unit is equipped with the Schwan S1073 ballast according to the schematic below. Note that the input power to the ballast is in series with the relay board.

[0059] POWER SWITCH: This switch is used to turn the device on and off, shutting off the overall power to the unit.

[0060] FAN: This fan is used for cooling inside the device and for cooling the treatment area. The specifications of the fan used are as follows:
Dimensions: 50 * 50 * 11.6 mm Voltage: 12dc Current: 0.08 amps

[0061] BOARD: The device board contains all the electronic components needed to operate the various parts. In the device board, the UV bulb and ballast in this board, are connected using standard wiring.

[0062] BOTTOM COVER: This frame is used to protect the components inside the device, which is made of plastic and its minimum and maximum operating temperatures range from minus 10 to 300 degrees Celsius. The device is sensitive to chloroform, dichloromethane, acetone, benzene, etc. It will be damaged if used. The parts placed on this frame are as follows:
UV lamp: It is attached to the lower frame with screws.
Ballast: It is attached to this frame after being placed on the stand.
The device board: It is attached to the top of the lower frame with a fixture.
The location of each part is shown on the exploded diagram.

[0063] GLASS TALC: This talc is used to allow the light from the lamp to pass through the device. It is made of a transparent polymer film that protects against ultraviolet rays and has a thickness of 2 mm.

[0064] AC SOCKET: It is used to connect the input power cord to the device.

[0065] COMB: This comb is used to push away the treatment area. It is made of PLA and printed by a 3D printer with an accuracy of 0.4 mm.

[0066] CAMERA: The device's camera consists of two parts, the 2-megapixel 0V2640 camera module and an E5P32 Wi-Fi module, and provides the ability to send images over a wireless network.
Date Recue/Date Received 2022-05-20

[0067] SHARP DISTANCE SENSOR: This sensor is used to measure the distance of the device to the treatment site (skin). The sensor model is gp2y0a21. Its specifications are as follows.
MIElectro-optical Characteristics_ I i--5V
Pararnelvr Svinboi t [ions 11.Y1. MAN.
I Itut A vcragQ= supply curd-criit I 0c11111Noty 31) mr%
Disimtce incusuruii.z Al. 1N4te I ) If) Kt) Lni )uiput vLt. r I 4i1t_iyi (NoTL' !) 0.4 1153 V

()wpm vokagc tlirftitx1/4:Ihti Ouiput voliage L9 2 17;
V
1. 1ocin mid 1 1-.(1ent Now. 1 9

[0068] BASE OF DISTANCE SENSOR: This base is used to place the distance sensor on the device to obtain a correct distance angle.

[0069] SCREW M3: It is used to hold the camera mount and sharp distance sensor.

[0070] SCREW M5: They are used to fast the overall container of the device.
They are actually the same 6 screws that hold the whole device.

[0071] CAMERA MOUNT: It is used to hold the camera on the handle and prevent the camera from falling.

[0072] CAMERA PAD: It is used to adjust the camera angle and position the camera completely parallel to the ground.

[0073] CELLPHONE HOLDER: It is designed to hold the phone on the handle. It makes easier for the users to use the phone and the UV device at the same time.

[0074] According to FIGS. lA to 1D, the distance between the device and the skin is measured by an embedded sensor. The intensity, dosage, and exposure time are estimated by the artificial intelligence algorithm developed for this device. Furthermore, the device has the maximum dimensions of 220 * 110 *
70 mm and weighs 0.8 0.1 kg, which allows the user to easily carry and operate it with one hand.
Device Software:

[0075] FIG. 2A and FIG. 2B are diagrams illustrating operation of a portable UVB device. According to these figures, a patient sits on a chair with his hand on a table. A portable UVB device is placed on the table and oriented to scan his hand. Once a scan is done, the data is sent to a doctor or technician for review and analysis. The distance between the device and the skin is measured by an embedded sensor in this device Date Recue/Date Received 2022-05-20 and the intensity, dosage, and exposure time are estimated by the artificial intelligence algorithm developed for this device.

[0076] FIG. 3 is a diagram illustrating an exemplary graphical user interface (GUI) for RULA analysis software. The UVB device software consists of the following:
Programming language: DartLang Implementing platform: Flutter Architecture: provider state management, mvc Libraries: tflite, flutter-blue, flutter-image IDE: Android studio

[0077] To use the UVB device, users place the device against the affected area. When it is turned on, they are asked to select their gender, age, skin type, skin color, and body parts they want to treat. Once the user's information and parameters of skin condition are measured and collected, the Al software processes the data and generates optimal phototherapy sessions uniquely tailored to each user's profile.

[0078] With the image processing ability, the software detects skin improvement after each session, thus self-adjusting UVB dosage and exposure time accordingly. When receiving the input from the distance sensor, the algorithm can direct the device to emit a suitable dose of UVB
rays and session duration based on the distance between the device and skin. The image processing system enables the device to optimize its treatment based on the progression of skin condition after every session.
Not only does this allow users to monitor their skin condition but it also adjusts the session if a lower dosage or shorter duration is needed.

[0079] Leveraging the Internet of Things (IoT) program, the AI-generated treatment schedule is sent to an integrated phone application with reminder notifications prior to and at treatment to ensure consistency and maximized effectiveness. The device is connected to a server storing electronic records of each user, allowing physicians to access their profiles for further diagnosis and follow-ups.

[0080] The device AI-based software can process input from the user's profile, sharp distance sensor, and camera to generate customized treatment schedules and sessions. Users can experience clinic-level of expertise at the push of the button. With the Al and IoT program, the device can automatically calibrate UVB dose and exposure time while keeping the users adhere to the treatment schedule.

Date Recue/Date Received 2022-05-20

[0081] The device is connected with a smartphone application via Bluetooth0.
The application displays relevant data and treatment schedule, as well as sending notifications prior to and at treatment time to ensure adherence.

[0082] The UVB device further includes a software application. The software application is used as follows:
= After installing the app, the splash page will appear and after a few seconds, the login page will appear. To enter the application, enter the username and password and click the login button. If the account is correct, the main page of the application will be accessed.
= The main page contains a four-part menu. The menus include the following:
Treatment, History, Notification, Profile.
= On the treatment page, you must first create the body area where you want to start treatment. To create the area (medical record), click the + button and fill out the form that appears.
= Elements to create a medical record: File name, Body area, Skin color, Doctor's name, Underlying diseases, Disease name.
= The disease name is recognized by the artificial intelligence (Al) network. After creating a medical record, it is sent to the doctor of choice and waits for the doctor to approve it. When the doctor views the record, the number of treatment sessions and treatment days required is determined.
= Click on the area the user created to start the treatment and click on the first active treatment session.
Before starting the treatment session, the user should select the skin color of the area where the previous session was irradiated and the number of days they were delayed in starting the session, and finally click the Start button. When the device is turned on, the application will connect to the device and send the start command to the device.
= The history page contains the history of treatment sessions. All notifications are placed on the notification page. The profile page contains user information and logging out of the user account.
Electronics and Printed Circuit Board (PCB):

[0083] FIG. 4 is a diagram illustrating an exemplary circuit board of a portable UVB device. According to FIG. 4, aspects of the circuit board include the follow components.

[0084] Input:
Input voltage: 5V
Input current: lA

Date Recue/Date Received 2022-05-20

[0085] Bluetooth0 Module: The device is equipped with the HCO5 Bluetooth0 SPP
(Serial Port Protocol) module, which is intended for establishing a transparent wireless serial connection. The Bluetooth0 serial port module is fully qualified Bluetooth0 V2.0+EDR (Enhanced Data Rate) 3Mbps Modulation with a complete 2.4GHz radio transceiver and baseband. It uses CSR Blue core 04-External single-chip Bluetooth0 system with CMOS technology and with AFH (Adaptive Frequency Hopping Feature).
Features of the Bluetooth module include the following:
= Typical -80dBm sensitivity = Up to +4dBm RF transmit power = Low power 1.8V operation,1.8 to 3.6V I/O
= PIO control = UART interface with programmable baud rate = With the integrated antenna = With edge connector

[0086] SD Card Module: This module is a Micro SD card reader module to read and write through the file system and SPI interface driver. SCM system can be completed inside a file Micro SD card. Features of the SC Card Module include:
= Support Micro SD card, Micro SDHC card (high-speed card).
= Communication interface is a standard SPI interface.
= Power supply is 4.5V ¨ 5.5V, 3.3V voltage regulator circuit board.

[0087] Microcontroller: The main computing element of the device is an Atmel0 ATmega128A
microcontroller. The Atmel0 ATmega128A is a low-power CMOS 8-bit microcontroller based on the AVRO enhanced RISC architecture. By executing powerful instructions in a single clock cycle, the ATmega128A achieves a throughput of nearly 'MIPS per MHz. This allows the system's designer to optimize the device in terms of power consumption and processing speed. Data measurements from sensors and Artificial Intelligence as well as interface control with the operator are the tasks performed by the MCU
in the device. Some of the key features of the device are listed below:
= High-performance, low-power Atmel AVR 8-bit microcontroller.
= Advanced RISC architecture.

Date Recue/Date Received 2022-05-20 = High endurance non-volatile memory segments.
= 128Kbytes of in-system self-programmable Flash program memory.
= 4Kbytes EEPROM.
= 4Kbytes internal SRAM.

[0088] FIG. 5 diagram illustrating a pre-trained MobileNet Deep CNN
classification system. According to FIG. 5, the pre-trained MobileNet Deep Convolutional Neural Network classification system with the help of Tensorflow0 framework and Keras library were used to classify the type of disease based on the image of psoriasis skin diseases.

[0089] In order to train the network, a suitable database was needed to drive the training process according to the samples. Considering the fact that data collection on medical topics is one of the most challenging issues. The best conditions for images of skin diseases are clinical data collection, but due to lack of time, it has been referred to the reference database. Com.Dermnet website as a complete reference of skin disease images, which includes 23 types of skin diseases, each of which contains training and evaluation data. In the network training process, the training data is used and the accuracy of the network is checked using evaluation data. In this data, there is a group related to psoriasis images which include 2000 color photos regarding different types of psoriasis. According to our requirements, 5 main groups were needed. Studies were conducted on psoriasis and its types, and based on the label of images, the images were manually divided into the 5 groups such as plaque, guttate, pustular, inverse, and erythrodermic for training and evaluation data. A total of 749 samples were obtained for training and 157 samples for evaluation.

[0090] To train the network, due to the small amount of data, the augmentation technique was used, which includes normalization, 90-degree rotation, horizontal and vertical shifts, etc., to strengthen the data. In the architecture used in this project, images with fixed dimensions are required for the input because the input images are of different sizes. First, all images were resized to 224 x 224 dimensions before training. Then input data in these dimensions were examined.

[0091] Due to the fact that the project is related to skin, it is also possible to use (Net Adversarial Generative work) by applying a variety of skin colors to increase and improve the use of data, which can be one of the future works.

[0092] Deep convolutional networks were used for training, which gives good results in image classification. The types of convolutional networks with a different number of convolutional layers were studied to extract the required features using Tensorflow0 and Keras library.
Pooling layers were used to Date Recue/Date Received 2022-05-20 reduce the increasing dimensions of convolutional. Dense layers were used for classification, and dropout was used to solve overfit problems. The next step was to examine some examples of pre-training networks such as mobilet, Resnet, Vgg, and Alexnet. Resnet50 showed the best performance. Better results were obtained by applying changes in the last layers. However, this model was discarded due to the high memory requirements for this architecture and the weights obtained. MobileNet was used due to the practicality of this project.

[0093] According to this disclosure, the existing weights of the network were not taken into account and retraining was performed. The last 5 layers were eliminated. 6 layers were implemented, including the Flatten, Dense, Dropout, Dense, Dropout, and Denset layers, which turned fine the model for the data project.

[0094] FIG. 6 is a diagram illustrating 86% Accuracy in Classifying Psoriasis Using the MobileNet Deep Neural Network and Augmentation. According to FIG. 6, network training using the optimization function RMSprop with a learning rate of 0001.0 and loss = categorical-crossentropy, performed with 140 IPAC, achieved a classification accuracy of 86% was used. Furthermore, deep neural network MobileNet and the augmentation, 5 types of psoriasis classified with 87% accuracy was also used.

[0095] FIG. 7 is a diagram illustrating a Confusion Matrix. According to FIG.
7, the accuracy of the model is indicated, which after 140 IPAC, the model achieved 95% accuracy with the education data and is shown in blue. For the evaluation data, which the model did not see any of this data during the training, with 86%
was able to recognize and classify the data and is shown shaded in the chart.

[0096] FIG. 7 also evaluates the efficiency of the model. The evaluation data is given to the model one by one and for each, one class is received as output. The predicted class and the actual data class can be displayed in a table.

[0097] FIG. 8A is a table illustrating various evaluation criteria for model testing. For example, out of 23 inverse data samples, 20 of them were found to be inverse, which is correct. 2 samples were found to be Plaque and one of them guttate. If the goal is to test the model for each of the classes, the table in FIG. 8A
is used, which contains various evaluation criteria.

[0098] To describe this better, consider the following case. FIG. 8B is a table illustrating two related outputs of each sample class. According to the table in FIG. 8B, a model that has two outputs, one positive and one negative. These two outputs actually represent the class of each sample.

Date Recue/Date Received 2022-05-20

[0099] When a model predicts a positive result, it wants to know how accurate that result is. If the value of false positives is high, accuracy is a good measure. For example, here the 84% accuracy for Plaque means that the model diagnosed the disease 84% correctly. If the value of false negatives is high, the recall criterion is a good one. The fl criterion is a good criterion for evaluating the accuracy of the model, which is "one"
in the best case and "zero" in the worst case.

[00100] This UV device has two applications. One for the client and the other for the doctor. FIG. 9 is a diagram illustrating Client Application overview. FIG. 10 is a diagram illustrating Doctor Application overview. According to FIG. 9 and FIG. 10, the UV applications are used to treat skin conditions with a UV device. The UV application works as follows.

[00101] After installing the application, the splash page will appear. After a few seconds, the login page will appear. To enter the application, the user should enter the username and password and then click on the login button. If the account is correct, the main page of the application will be accessed. The main page contains the following menus. The menus include the Treatment and Profile pages.

[00102] The treatment page can display the treatment records that the user has already created. By clicking on each record, one can see the details of each record and continue the treatment process. When the user wants to treat an area of the body, s/he creates a treatment record. Then the user clicks the + button and fills out the form that appears. The items in the medical record form include the following: Record name, Body area, Skin color, Doctor name, Underlying diseases, Disease name, and Area size.

[00103] The disease name is recognized by the artificial intelligence network.
To diagnose the disease name, click on the "Add Disease" button and the user will first be asked for the type of treatment. The treatment options include corona and psoriasis options. After selecting the treatment type, if the user is connected to the device camera, the camera image will be displayed, and clicking the Al detection button will diagnose the skin disease.

[00104] After a medical record is created, the record is sent to the doctor of the user's choice and awaits the doctor's approval. When the doctor reviews the record, the number of treatment sessions and treatment days required are determined.

[00105] To start the treatment, the user clicks on the created record then clicks on the active treatment session. Before starting the treatment session, one should select the skin color of the area where the previous session was irradiated. After that, the user clicks the "Calculate Dosage"
button. At this moment, the phone is connected to the device. It calculates the required dose based on the user information and displays it on Date Recue/Date Received 2022-05-20 the screen. Then, clicking on the "Start" button, the command to start the treatment process is sent to the device. The device begins to work.

[00106] After the end of the treatment process, the user clicks on the treatment session and records the end of it. At the beginning of the treatment, if the user selects the skin color of the previous session red, the record will be sent to the doctor. Then the session will need the doctor's approval to start the treatment process.

[00107] On the record page of the list of treatment sessions along with the percentage of progress, the start date of treatment can also be seen. The profile page contains the user information and logout.

[00108] The diagnosis is created by the deep learning artificial intelligence network in the Python language and the Tensorflow0 Lite library. The way the network works is using the IP
camera. The skin environment is photographed. The image frame is resized in the dimensions of 256 * 256 and stored on the internal memory of the phone, then the network is loaded in the app. After that is sent to the network. The name of the disease is visible in the network output.

[00109] The UV device is used to treat the disease. In the application, when creating a treatment record, the necessary information for using the device is taken from the user. When the treatment starts, this information is sent to the device via Bluetooth0 connection and the USART
protocol. After receiving the disease information, the device calculates the radiation dose and its working time, then it starts irradiating the radiation on the skin.

[00110] Medical records and treatment sessions are stored on the server using web services. The doctor in charge of the record can monitor the patient's recovery. The user (patient) creates a medical record through the patient application and the record information is sent to the doctor application and the physician can visit the patient record.

[00111] To use the application for doctors, first the login page is opened, the doctor enters his user information and gets to the main page of the application. On the main page there are two tabs, the "Folders"
tab and the "Treatments" tab.

[00112] The Folder tab displays the files referred to the doctor, and the doctor can view the details of each file by selecting it and then confirming the file by entering the treatment start date, number of sessions, and time between sessions.

Date Recue/Date Received 2022-05-20

[00113] The Treatment tab can display the treatment sessions of all patients that need to be confirmed by the doctor. The doctor can confirm the session after seeing the details of the session.

[00114] Records are created by the user application and then stored on the server database using the rest API. The files are visible in the doctor application.
Treatment Dosage:

[00115] FIG. 8C is a table illustrating appropriate dose for each skin color to treat psoriasis. FIG. 8D is a table illustrating appropriate dose to use in corona. According to FIG. 8C and FIG. 8D, the parameters affecting the dose are as follows:

[00116] Treatment: For treatment of psoriasis or coronavirus.

[00117] Patient's skin color: It is divided into 6 general categories from white (pale) to black. The appropriate dose for each skin color for the treatment of psoriasis is shown in tables in FIG. 8C and FIG.
8D.

[00118] Body Area: the choice of body area is important because the maximum dose for the face in any treatment model (corona or psoriasis) is equal to 1000 mJ/cm^2. In other words, the above tables are not correct for the face.

[00119] Number of Treatment Sessions: Such that the patient requires a total of 40 treatment sessions.
Therefore, the dose used in each session varies from the minimum dose to the maximum dose allowed in relation to the number of sessions.
Dosage = (maxdose-mindose) * (current session / total sessions) + mindose.

[00120] For example, if the treatment session is 20 and the minimum and maximum doses are 0.5 and 1 watt/cm, respectively, the dose is calculated as follows:
Dosage = (1-0.5) * (20/40) + 0.5 = 0.75

[00121] Correction Rate for Delayed Sessions: If for any reason the patient has been unable to continue treatment continuously, the dose should be adjusted as follows:
1-week treatment interruption: the last dose should be repeated. In other words, the pre-session will be repeated.

Date Recue/Date Received 2022-05-20 1 to 2 weeks: The previous dose should be reduced by 25%.
2 to 4 weeks: The previous dose should be reduced by 50%.
More than 4 weeks: Treatment should be restarted from the beginning.

[00122] Skin Inflammation: If the skin becomes inflamed and irritated under radiation treatment, the dose of the previous session should be applied without an increase. If the inflammation is very severe, the application will automatically notify the doctor and the treatment will be stopped.

[00123] In general, taking into account all the above parameters, the formula for calculating the dose in the device will be as follows:
if (region == 0)! * face * /
maxdose = 1.00;
else / * other body regions * /
maxdose = DOSEMAX [curetype] [skincolor];
Dosage = (((maxdose - DOSEMIN [curetype] [skincolorp* (float) (session-erythema- (correction>
0)) / 39.0) + DOSEMIN [curetype] [skincolor]) * ((100.0- (float) correction) /100.0);

[00124] The maximum and minimum values are defined as follows (in J/cm^2):
float DOSEMAX [2] [6] = {{1.70,1.700,2.500,2.500,3.400,3.400}, / * corona * /
{2.000,2.000,3.000,3.000,5.000,5.000}1; / * for psoriasis * /
float DOSEMIN [2] [6] = 11.300, .300, .500, .500, .800, .8001, / * corona * /
{.300, .300, .500, .500, .800, .800}1; / * for psoriasis * /

[00125] UV Bluetooth Connection: The device transmits the following string once every two seconds:
# UV **
After selecting the automatic mode, the parameters for color, area, etc.
should be sent from the phone as follows:
#cmd cure = 0 skin = 3 regi = 1 dise = 0 sess = 1 con = 0 eryt = 0 size = 0 **
Where the letters cmd indicate the command sent to the device and cure: Curetype: corona treatment 0 or psoriasis 1 Date Recue/Date Received 2022-05-20 skin: skincolor: skin color (number between 0 and 5) regi: region: (number between 0 and 4) body area dise: disease: (number between 0 and 5) type of disease sess: session: multiple sessions (number between 1 and 40) con: correction: correction factor for forgotten sessions (number between 0 and 100) erythema: erythema (number between 0 and 2) size: areasize: has no effect on the formula

[00126] Once the above parameters are received, the dose is calculated and in addition to the daily display LCD, the following device string is sent to the phone containing the calculated dose to be displayed to the user of the application.
# dosage = 1.61 **
The dose number is a decimal number with two decimal places.
The device then begins to operate by receiving the following confirmation string from the phone:
# confirm **

[00127] Implementations disclosed herein provide systems, methods and apparatus for generating or augmenting training data sets for machine learning training. The functions described herein may be stored as one or more instructions on a processor-readable or computer-readable medium. The term "computer-readable medium" refers to any available medium that can be accessed by a computer or processor. By way of example, and not limitation, such a medium may comprise RAM, ROM, EEPROM, flash memory, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer. It should be noted that a computer-readable medium may be tangible and non-transitory. As used herein, the term "code" may refer to software, instructions, code or data that is/are executable by a computing device or processor. A "module" can be considered as a processor executing computer-readable code.

[00128] A processor as described herein can be a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A
general purpose processor can Date Recue/Date Received 2022-05-20 be a microprocessor, but in the alternative, the processor can be a controller, or microcontroller, combinations of the same, or the like. A processor can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor may also include primarily analog components. For example, any of the signal processing algorithms described herein may be implemented in analog circuitry. In some embodiments, a processor can be a graphics processing unit (GPU). The parallel processing capabilities of GPUs can reduce the amount of time for training and using neural networks (and other machine learning models) compared to central processing units (CPUs). In some embodiments, a processor can be an ASIC including dedicated machine learning circuitry custom-build for one or both of model training and model inference.

[00129] The disclosed or illustrated tasks can be distributed across multiple processors or computing devices of a computer system, including computing devices that are geographically distributed. The methods disclosed herein comprise one or more steps or actions for achieving the described method. The method steps and/or actions may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is required for proper operation of the method that is being described, the order and/or use of specific steps and/or actions may be modified without departing from the scope of the claims.

[00130] As used herein, the term "plurality" denotes two or more. For example, a plurality of components indicates two or more components. The tenn "determining" encompasses a wide variety of actions and, therefore, "determining" can include calculating, computing, processing, deriving, investigating, looking up (e.g., looking up in a table, a database or another data structure), ascertaining and the like. Also, "determining" can include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory) and the like.

[00131] The phrase "based on" does not mean "based only on," unless expressly specified otherwise. In other words, the phrase "based on" describes both "based only on" and "based at least on." While the foregoing written description of the system enables one of ordinary skill to make and use what is considered presently to be the best mode thereof, those of ordinary skill will understand and appreciate the existence of variations, combinations, and equivalents of the specific embodiment, method, and examples herein. The system should therefore not be limited by the above described embodiment, method, and examples, but by all embodiments and methods within the scope and spirit of the system. Thus, the present disclosure is not intended to be limited to the implementations shown herein but is to be accorded the widest scope consistent Date Recue/Date Received 2022-05-20 with the principles and novel features disclosed herein.

Date Recue/Date Received 2022-05-20 SYSTEM AND METHOD FOR UTILIZING A PORTABLE PHOTOTHERAPY DEVICE FOR
THE TREATMENT OF PSORIASIS
Cross Reference to Related Applications [0001] The application claims priority to and the benefit of US Provisional Patent Application Serial No.
63/191056, entitled "SYSTEM AND METHOD FOR UTILIZING A PORTABLE PHOTOTHERAPY
DEVICE FOR THE TREATMENT OF PSORIASIS" filed on May 20, 2021.
Background [0002] The embodiments described herein relate to phototherapy technology for treating psoriasis.
[0003] Psoriasis is one of the most common human skin diseases and appears to have an important genetic basis. It is a chronic, healing, recurrent scaling skin disease that affects 1 to 3 percent of the world's population. The diagnosis is made on a clinical basis, although histological examination of the skin biopsy specimen may be helpful. Psoriasis is a debilitating, though rarely life-threatening, disease with social and economic implications that is underappreciated by physicians and other health care providers. Recently, there has been progress in understanding the pathogenesis of psoriasis, and therapeutic advances have improved the care of even the most severely injured patients.
[0004] Psoriasis is a common, chronic, non-communicable skin disease with no clear cause or cure, causing red, itchy, scaly patches. Around 125 million individuals are affected worldwide, making psoriasis a serious global problem with a reported prevalence ranging between 0.09% and 11.43%. According to the WHO Global report on psoriasis, many people suffer from this skin disease due to delayed diagnosis, inadequate treatment, and insufficient access to care. There is an upward trend in the prevalence of psoriasis across countries. Data from the National Health and Nutrition Examination Survey indicated an increase in prevalence from 1.62% in 2004 to 3.10% by 2010 in the United States. The rising prevalence of psoriasis drives the need for psoriasis treatments.
[0005] Fortunately, psoriasis is characterized by overgrowth and abnormal keratinocyte differentiation, but it is completely reversible with proper treatment. The trigger for keratinocyte responses is thought to be activation of the cellular immune system; T cells, dendritic cells, and immune-related cytokines and chemokines are involved in pathogenesis.
[0006] Psoriasis may get worse with age. The degree of severity depends on hereditary and environmental Date Recue/Date Received 2022-05-20 factors. Its severity ranges from a few red plaques and scattered scales to involvement of almost the entire surface of the body. Recent data shows that the incidence of psoriasis varies according to geographical area;
it is more common in countries far from the equator. Studies on the prevalence and incidence of psoriasis have helped to better understand the burden of the disease.
[0007] Psoriasis can be treated by using various topical and biologic drug treatments. But each has its own properties or limitations. For example, pills, ointments, salicylic acid, bandages, etc., can stop the disease but not cure it. There are other options like oral or injected drugs, skin grafting or genetic modification, and everything related to it, which is used for the treatment of skin diseases.
[0008] There are different types of treatments used for controlling the symptoms, such as topical and systemic therapies as well as phototherapy. Each method has its pros and cons.
However, topical and systemic therapies are known to have a wide array of unintended side effects and immunosuppression.
Furthermore, the global market for psoriasis drugs is declining at a CAGR of -10.8% due to COVID-19's lockdown measures and restrictions of producing non-emergency medicine. Such a trend drives more demand toward alternative treatment methods such as hand-held phototherapy devices that can be administered at home and produced without many medical resources. As such, the preferred method of treatment nowadays is phototherapy with the use of narrowband Ultraviolet B
(NB-UVB) radiation rays.
[0009] It can be practiced in clinics with specialized phototherapy machines, requiring two to three applications per week. However, busy schedules make it hard for patients to balance their 9-5 workdays with their mandatory phototherapy sessions occurring multiple times a week at a clinic. Moreover, due to COVID-19 pandemic, patients are prohibited from in-person doctor visits for non-urgent diseases in order to limit the risk of developing or spreading the virus. Such instances lead to the development of a home-based phototherapy device that users can apply at their convenience. However, the self-application of UVB
rays is susceptible to human errors. Improper and infrequent use can increase the risk of skin cancer, skin thinning, skin burns, and treatment failure.
[0010] Light therapy is a first-line treatment for moderate to severe psoriasis, either alone or in combination with medication. It involves exposing the skin to a controlled amount of natural or artificial light. Repeated treatments are necessary. Therefore, the patient can talk to his / her doctor about whether home phototherapy is an option for him / her.
[0011] There is a desire to provide a safe, reliable, and cost-effective home-based phototherapy device to help patients treat psoriasis. There is a desire for a portable phototherapy device, using Ultraviolet B (UVB) radiation, for the treatment of psoriasis. There is a further desire to use artificial intelligence to assist in the Date Recue/Date Received 2022-05-20 diagnosis of different types of psoriasis and intelligently control the dosage rate.
Summary [0012] The present innovation is a newer portable type of narrow-band ultraviolet-B (UVB) device, especially a type of targeted phototherapy in terms of medical and cosmetic treatment. The UVB device is an apparatus with a new approach to detecting and treating some skin diseases such as different types of psoriasis (plaque, Pustular, flexural, Guttate, and Erythroderma), Vitiligo, Pityriasis Rosea, Atopic eczema, Hyper bilirubin, and other diseases on different skin types. One of the advantages and technological parts of this device is image processing. When a person is treated, his skin gradually improves, and the image processing shows this improvement, so a lower dose is needed. All devices developed for the treatment of psoriasis have the same function and do not differ from each other.
[0013] This device includes artificial intelligence to diagnose five types of psoriasis and intelligent control for dosage rate. This device also has image processing that can be helpful to show the improvement of the skin. It should be noted that ultraviolet B (UVB) radiation increases cutaneous vitamin D synthesis.
Therefore, treatment with the above device leads to an increase in vitamin D
absorption which can help boost the immune system to fight other diseases and viruses (e.g., coronavirus).
[0014] Furthermore, this device has a combination of both hardware and software. It includes a sharp distance sensor, a camera for image processing, a Bluetooth module for wireless connectivity and an IoT
program. The camera takes pictures of a person's skin, then the severity of the disease is determined through it.
Brief Description of the Drawings [0015] FIG. IA is an exploded view of an assembly of a portable UVB device.
[0016] FIG. 1B is a table listing different component of a portable UVB
device.
[0017] FIG. IC is aback perspective view of a portable UVB device.
[0018] FIG. ID is a front perspective view of a portable UVB device.
[0019] FIG. 2A and 2B are diagrams illustrating operation of a portable UVB
device.
[0020] FIG. 3 is a diagram illustrating an exemplary graphical user interface (GUI) for RULA analysis software.

Date Recue/Date Received 2022-05-20 [0021] FIG. 4 is a diagram illustrating an exemplary circuit board of a portable UVB device.
[0022] FIG. 5 diagram illustrating a pre-trained MobileNet Deep CNN
classification system.
[0023] FIG. 6 diagram illustrating 86% Accuracy in Classifying Psoriasis Using the MobileNet Deep Neural Network and Augmentation.
[0024] FIG. 7 is a diagram illustrating a Confusion Matrix.
[0025] FIG. 8A is a table illustrating various evaluation criteria for model testing.
[0026] FIG. 8B is a table illustrating two related outputs of each sample class.
[0027] FIG. 8C is a table illustrating appropriate dose for each skin color to treat psoriasis.
[0028] FIG. 8D is a table illustrating appropriate dose to use in corona.
[0029] FIG. 9 is a diagram illustrating Client Application overview.
[0030] FIG. 10 is a diagram illustrating Doctor Application overview.
Detailed Description [0031] Psoriasis is a chronic and noncontagious skin disease, which produces red, itchy, and raised patches on the skin. Psoriasis can be treated by applying various treatments including topical therapies, biologic drug treatments, systemic therapies and hospital-based phototherapy. Each has its own characteristics or limitation.
Psoriasis Treatment Options:
[0032] Topical therapies, including creams and ointments applied on top of the affected skin, are popular alternatives for mild cases as they are effective in treating symptoms.
However, they are not a substitute for the treatment that healthcare providers recommend. One major drawback regarding topical treatments is that many contain chemicals, fragrances, and dyes that can irritate the skin and create discomfort or a worsening of symptoms. Additional side effects can include an increased risk for skin cancer, lymphoma, and increased sensitivity to light.

Date Recue/Date Received 2022-05-20 [0033] Systemic therapies include the use of biologics and are taken orally or through injections. This includes drugs such as Remicade , Humira , and Enbrel , among others. Systemic therapies are also not recommended for patients who find themselves immunocompromised, pregnant, or breastfeeding.
[0034] The use of biologic drug treatments and injections, is only prescribed for brief periods of time, often alternating with other treatments due to their severe side effect, namely osteoporosis, cancer, digestive and kidney issues. Another factor to note about biologics is that patients are seeing an increase in out-of-pocket costs associated with the pharmaceutical purchases of drugs as some insurance companies have revoked their deductible status, costing patients thousands of dollars.
[0035] Receiving phototherapy treatment at a clinic is a safe yet costly and inconvenient alternative.
Including travel expenses and lost wages, three months of in-office UVB
phototherapy can cost over $1000, making it more expensive per treatment than home-based phototherapy.
Furthermore, clinical-based treatments have waiting times associated with seeing medical specialists in certain countries.
Phototherapy as a Recommended Treatment:
[0036] From all psoriasis treatments available, phototherapy remains the safest option, according to the National Psoriasis Foundation, due to having fewer side effects than biologics. According to the National Psoriasis Foundation, biologics have a chance of increasing infections, which are then required to be monitored to prevent any adverse effects from occurring. Not only is phototherapy safe, but it is also considered to be as effective or more effective than other treatments, according to a published article in the Journal of the American Academy of Dermatology.
[0037] The main differentiators when it comes to home-based and hospital-based phototherapy are cost and accessibility. Home-based units are a one-time purchase that can save patients time, money, and the hassle of frequent clinic visits. Home-based solutions allow patients to manage their schedules without stress as their weekly treatments can be conducted in the comfort of their own home at their self-scheduled times, avoiding conflict with other commitments such as work.
[0038] Patients also avoid the extremely long wait times associated with receiving treatment in Canada, enabling them to consistently treat their condition. In 2013, an economic analysis also concluded that purchasing a home-based phototherapy solution was more cost-effective for patients living 20 miles or more from a clinic, regardless of whether the device was covered by insurance.
Date Recue/Date Received 2022-05-20 Ultraviolet B (UVB) Phototherapy Device:
[0039] An Ultraviolet B (UVB) phototherapy device is an apparatus with a new approach for detecting and treatment of skin disorders, such as Psoriasis (Plaque, Pustular, Flexural, Guttate, and Erythrodermic), Vitiligo, Pityriasis rosea, Atopic eczema, Hyper bilirubin and other diseases on all skin types.
[0040] Ultraviolet (UV) Phototherapy is a method to prevent and cure diseases using artificial ultraviolet radiation. Devices with ultraviolet light function have been developed to treat psoriasis. However, due to their special functionality, these devices require a skilled operator.
Furthermore, they are difficult to use.
Existing devices developed to treat psoriasis have the same function and are not different from each other.
[0041] The present innovation is a newer portable type of narrow-band ultraviolet-B (UVB) device, especially a type of targeted phototherapy in terms of medical and cosmetic treatment. The UVB device is an apparatus with a new approach to detecting and treating some skin diseases, such as psoriasis (plaque, Pustular, flexural, Guttate, and Erythroderma), Vitiligo, Pityriasis Rosea, Atopic eczema, Hyper bilirubin, and other diseases on all skin types.
[0042] One of the advantages and technological parts of this device is image processing. When a person is treated, his skin gradually improves and the image processing shows this improvement, so a lower dose is needed. All devices developed for the treatment of psoriasis have the same function and do not differ from each other.
[0043] The present innovation is a newer portable type of ultraviolet B (UVB) device called narrowband UVB (NB-UVB), especially a kind of targeted phototherapy, one which is in contact with medical and beauty treatment. Therefore, the targeted customers are everyone who has a skin disease, including individuals, beauty clinics and hospitals.
[0044] The narrowband UVB produces the part of the ultraviolet light spectrum band that can be helpful for psoriasis. The lamp of UVB psoriasis using that feature can affect the skin on the cellular level. UVB
light is radiated in a way that can solve the mentioned problem. The UV
radiation causes the cells to mutate and then incomplete metabolism. Therefore, this disease can be treated by killing cells.
[0045] The device has a combination of both hardware and software. This includes a sharp distance sensor, a camera for image processing, and IoT program. To use it, it can be easily placed against the affected area.
When the device is turned on, the user is asked to select their skin type or its color. It also asks which part of body he / she wants to treat. The device has the ability to ask the user how many sessions of treatment it Date Recue/Date Received 2022-05-20 is. On the other hand, it has a Bluetooth module and a camera. The camera takes pictures of a person's skin, then determines the severity of the disease through image processing. In the last step, all the information is entered into the table in order to make a decision about a suitable dose of UVB ray that should radiate on a person's skin. The distance of the skin from the device is constantly being measured. If a person moves his/her hand, the sharp distance sensor constantly senses the distance.
[0046] The mentioned device can be used for treating all parts of the body, such as the scalp. The scalp is one of the most difficult areas to treat because the hair hinders the direct effect of ultraviolet irradiation. It has a special comb attachment that can also be used as a distance guard to treat other parts. Consequently, in the device, four parameters are examined: type of disease, skin type, severity of the disease and treatment session. The obtained results are in accordance with the parameters that dermatologists use as a standard criterion for the diagnosis of psoriasis. It can be noted that this portable device has 2 combs on each side of it for putting aside hair.
[0047] This device includes artificial intelligence (Al) to diagnose up to five types of psoriasis and intelligent control for dosage rate. This device also has image processing that can be helpful to show the improvement of the skin.
[0048] Additionally, there is a touched LCD for showing the results and whatever is related. In fact, one of the advantages and technological parts of this device is image processing.
When a person is treated, his skin gradually improves and the image processing shows this improvement, so a lower dose is needed.
[0049] It should be noted that ultraviolet B (UVB) radiation increases cutaneous vitamin D synthesis.
Therefore, treatment with the above device leads to an increase in vitamin D
absorption. Obviously, vitamin D can be helpful in boosting your immune system to fight coronavirus.
Device Hardware:
[0050] The device is a hand-held device equipped with a UVB lamp, a sharp-distance sensor, a camera for image processing and an LCD touch screen. The device emits UVB light that can be used to treat individual parts of the user's body, reducing inflammation and slowing the production of skin cells. This helps manage and prevent a psoriasis flare. The sharp-distance sensor measures the distance of the skin from the device, allowing it to automatically adjust the dosage and exposure time based on your motions. This feature reduces the risk of skin burns, freckles formation, and increased sun sensitivity associated with overexposure to UVB narrowband. The camera will take pictures of the affected skin, determining four parameters: type of disease, skin type, the severity of the disease, and treatment session via image Date Recue/Date Received 2022-05-20 processing. Once the device completes the diagnosis, it will show results and relevant user instructions on the LCD touch screen. The device also comes with two comb attachments used for treating the scalp area and protecting unaffected skin from exposure to UVB rays.
[0051] FIGS. IA to ID are diagrams illustrating different parts of a portable UVB device assembly including a top cover, LCD, UV protector, ballast base, UV lamp, ballast, power switch, fan, board, lower cover, glass talc, ac socket, comb, camera, distance sensor, and the base of the distance sensor. According to these figures, FIG. IA is an exploded view of an assembly of a portable UVB
device. FIG. 1B is a table listing different components of a portable UVB device. FIG. IC is a back perspective view of a portable UVB device. FIG. ID is a front perspective view of a portable UVB device.
[0052] This device has the maximum dimensions of 220 mm (length) X 110 mm (width) X 70 mm (depth) and weighs 0.8 0.1 kg, which allows the user to easily carry and operate it with one hand. According to FIGS. IA and 1B, further details of these components are as follows.
[0053] TOP COVER: This frame is used to protect the components inside the device, which is made of plastic and its minimum and maximum operating temperatures range from minus 10 to 300 degrees Celsius.
The device is sensitive to chloroform, dichloromethane, acetone, benzene, etc.
It will be damaged if used.
[0054] LCD: It is a 3.2-inch color screen that displays device status and treatment progress, and is equipped with a resistive touch screen. The screen resolution is 320 x 240 pixels and 64 thousand colors.
The following details show the characteristics of the LCD:
= 3.2-inch color screen, support 16BIT RGB 65K color display, displays rich colors.
= 240x320 resolution for clear display.
= Support 8-bit/16-bit parallel bus switching (default is 16-bit), fast transmission speed.
= Supports ALIENTEK 5TM32 Mini, Elite, Warship, Explorer, and Apollo development boards for direct plug-in use.
= Supports touch function.
= Supports SD card function expansion.
= Provides a comprehensive sample program for 5TM32 and C51 platforms.
= Military process standards, long-term stability.
= Provides underlying technical support for drivers.
[0055] UV PROTECTOR: The UV Protector is designed to protect the lamp and prevent light radiation to the internal components, as well as to use the maximum power of the lamp for the treatment.
[0056] BALLAST BASE: It is used to hold the ballast in the unit.

Date Recue/Date Received 2022-05-20 [0057] UV LAMP: This lamp is type c and is used for treatment. The UV lamp installed in device is PL -S 9W/01/2P manufactured by Philips . The UV lamp has biological safety - IEC
62471 - Risk Group 3.
The following details show the characteristics of the UVB lamp:
1. General Information = Cap-Base: G23 [ G23]
= Life to 50% Failures (Nom): 1000 h = Useful Life (Nom): 1000 h 2. Light Technical = Color Code: 01 = Color Designation: Ultra Violet B
= UV Depreciation at 500 h: 15%
= UV Depreciation at 1000 h: 20%
3. Operating Electrical = Power (Nom): 8.6 W
= Lamp Current (Nom): 0.17 A
= Voltage (Nom): 60 V
4. Mechanical and Housing = Cap-Base Information: 2 Pins 5. Approval and Application = Mercury (Hg) Content (Nom): 3.0 mg 6. UV
= UV-B Radiation 100 hr. (IEC): 1.0 w = UV-B Radiation 5hr (IEC): 1.2 W
[0058] BALLAST: In a fluorescent lighting system, the ballast controls the current to the lamps and provides sufficient voltage to start the lamps. Without a ballast to limit the current, a fluorescent lamp connected directly to a high voltage source would increase its current draw rapidly and uncontrollably.
Within a second, the lamp would overheat and burn out. During lamp start-up, the ballast must supply high voltage for a short time to create an arc between the two lamp electrodes.
Once the arc is established, the ballast quickly reduces the voltage and regulates the electrical current to produce a uniform light output.

Date Recue/Date Received 2022-05-20 The unit is equipped with the Schwan S1073 ballast according to the schematic below. Note that the input power to the ballast is in series with the relay board.
[0059] POWER SWITCH: This switch is used to turn the device on and off, shutting off the overall power to the unit.
[0060] FAN: This fan is used for cooling inside the device and for cooling the treatment area. The specifications of the fan used are as follows:
Dimensions: 50 * 50 * 11.6 mm Voltage: 12dc Current: 0.08 amps [0061] BOARD: The device board contains all the electronic components needed to operate the various parts. In the device board, the UV bulb and ballast in this board, are connected using standard wiring.
[0062] BOTTOM COVER: This frame is used to protect the components inside the device, which is made of plastic and its minimum and maximum operating temperatures range from minus 10 to 300 degrees Celsius. The device is sensitive to chloroform, dichloromethane, acetone, benzene, etc. It will be damaged if used. The parts placed on this frame are as follows:
UV lamp: It is attached to the lower frame with screws.
Ballast: It is attached to this frame after being placed on the stand.
The device board: It is attached to the top of the lower frame with a fixture.
The location of each part is shown on the exploded diagram.
[0063] GLASS TALC: This talc is used to allow the light from the lamp to pass through the device. It is made of a transparent polymer film that protects against ultraviolet rays and has a thickness of 2 mm.
[0064] AC SOCKET: It is used to connect the input power cord to the device.
[0065] COMB: This comb is used to push away the treatment area. It is made of PLA and printed by a 3D printer with an accuracy of 0.4 mm.
[0066] CAMERA: The device's camera consists of two parts, the 2-megapixel 0V2640 camera module and an E5P32 Wi-Fi module, and provides the ability to send images over a wireless network.
Date Recue/Date Received 2022-05-20 [0067] SHARP DISTANCE SENSOR: This sensor is used to measure the distance of the device to the treatment site (skin). The sensor model is gp2y0a21. Its specifications are as follows.
=Electro-optical Characteristics n ,\ 5V) l'aratua:r Symbol 'ontl it ions miN vp, nitAx. !Au A vcrage. supplv carrunr J140011 N OIC 30 40 111A_ tuL=a,L11 \ iNotc I ) if) SO cm Our )Lit V (Nuto I ) 0.25 OA
OuLjull drag,: tliFfuo..): 11):1\N.L\
Wpm vOlatp: 641/0 11.65 L9 2.15 L ;Ind I fle:11(Nolo ) [0068] BASE OF DISTANCE SENSOR: This base is used to place the distance sensor on the device to obtain a correct distance angle.
[0069] SCREW M3: It is used to hold the camera mount and sharp distance sensor.
[0070] SCREW MS: They are used to fast the overall container of the device.
They are actually the same 6 screws that hold the whole device.
[0071] CAMERA MOUNT: It is used to hold the camera on the handle and prevent the camera from falling.
[0072] CAMERA PAD: It is used to adjust the camera angle and position the camera completely parallel to the ground.
[0073] CELLPHONE HOLDER: It is designed to hold the phone on the handle. It makes easier for the users to use the phone and the UV device at the same time.
[0074] According to FIGS. lA to 1D, the distance between the device and the skin is measured by an embedded sensor. The intensity, dosage, and exposure time are estimated by the artificial intelligence algorithm developed for this device. Furthermore, the device has the maximum dimensions of 220 * 110 *
70 mm and weighs 0.8 0.1 kg, which allows the user to easily carry and operate it with one hand.
Device Software:
[0075] FIG. 2A and FIG. 2B are diagrams illustrating operation of a portable UVB device. According to these figures, a patient sits on a chair with his hand on a table. A portable UVB device is placed on the table and oriented to scan his hand. Once a scan is done, the data is sent to a doctor or technician for review and analysis. The distance between the device and the skin is measured by an embedded sensor in this device Date Recue/Date Received 2022-05-20 and the intensity, dosage, and exposure time are estimated by the artificial intelligence algorithm developed for this device.
[0076] FIG. 3 is a diagram illustrating an exemplary graphical user interface (GUI) for RULA analysis software. The UVB device software consists of the following:
Programming language: DartLang Implementing platform: Flutter Architecture: provider state management, mvc Libraries: tflite, flutter-blue, flutter-image IDE: Android studio [0077] To use the UVB device, users place the device against the affected area. When it is turned on, they are asked to select their gender, age, skin type, skin color, and body parts they want to treat. Once the user's information and parameters of skin condition are measured and collected, the Al software processes the data and generates optimal phototherapy sessions uniquely tailored to each user's profile.
[0078] With the image processing ability, the software detects skin improvement after each session, thus self-adjusting UVB dosage and exposure time accordingly. When receiving the input from the distance sensor, the algorithm can direct the device to emit a suitable dose of UVB
rays and session duration based on the distance between the device and skin. The image processing system enables the device to optimize its treatment based on the progression of skin condition after every session.
Not only does this allow users to monitor their skin condition but it also adjusts the session if a lower dosage or shorter duration is needed.
[0079] Leveraging the Internet of Things (IoT) program, the AI-generated treatment schedule is sent to an integrated phone application with reminder notifications prior to and at treatment to ensure consistency and maximized effectiveness. The device is connected to a server storing electronic records of each user, allowing physicians to access their profiles for further diagnosis and follow-ups.
[0080] The device AI-based software can process input from the user's profile, sharp distance sensor, and camera to generate customized treatment schedules and sessions. Users can experience clinic-level of expertise at the push of the button. With the Al and IoT program, the device can automatically calibrate UVB dose and exposure time while keeping the users adhere to the treatment schedule.

Date Recue/Date Received 2022-05-20 [0081] The device is connected with a smartphone application via Bluetooth .
The application displays relevant data and treatment schedule, as well as sending notifications prior to and at treatment time to ensure adherence.
[0082] The UVB device further includes a software application. The software application is used as follows:
= After installing the app, the splash page will appear and after a few seconds, the login page will appear. To enter the application, enter the username and password and click the login button. If the account is correct, the main page of the application will be accessed.
= The main page contains a four-part menu. The menus include the following:
Treatment, History, Notification, Profile.
= On the treatment page, you must first create the body area where you want to start treatment. To create the area (medical record), click the + button and fill out the form that appears.
= Elements to create a medical record: File name, Body area, Skin color, Doctor's name, Underlying diseases, Disease name.
= The disease name is recognized by the artificial intelligence (Al) network. After creating a medical record, it is sent to the doctor of choice and waits for the doctor to approve it. When the doctor views the record, the number of treatment sessions and treatment days required is determined.
= Click on the area the user created to start the treatment and click on the first active treatment session.
Before starting the treatment session, the user should select the skin color of the area where the previous session was irradiated and the number of days they were delayed in starting the session, and finally click the Start button. When the device is turned on, the application will connect to the device and send the start command to the device.
= The history page contains the history of treatment sessions. All notifications are placed on the notification page. The profile page contains user information and logging out of the user account.
Electronics and Printed Circuit Board (PCB):
[0083] FIG. 4 is a diagram illustrating an exemplary circuit board of a portable UVB device. According to FIG. 4, aspects of the circuit board include the follow components.
[0084] Input:
Input voltage: 5V
Input current: IA

Date Recue/Date Received 2022-05-20 [0085] Bluetooth Module: The device is equipped with the HCO5 Bluetooth SPP
(Serial Port Protocol) module, which is intended for establishing a transparent wireless serial connection. The Bluetooth serial port module is fully qualified Bluetooth V2.0+EDR (Enhanced Data Rate) 3Mbps Modulation with a complete 2.4GHz radio transceiver and baseband. It uses CSR Blue core 04-External single-chip Bluetooth system with CMOS technology and with AFH (Adaptive Frequency Hopping Feature).
Features of the Bluetooth module include the following:
= Typical -80dBm sensitivity = Up to +4dBm RF transmit power = Low power 1.8V operation,1.8 to 3.6V I/O
= PIO control = UART interface with programmable baud rate = With the integrated antenna = With edge connector [0086] SD Card Module: This module is a Micro SD card reader module to read and write through the file system and SPI interface driver. SCM system can be completed inside a file Micro SD card. Features of the SC Card Module include:
= Support Micro SD card, Micro SDHC card (high-speed card).
= Communication interface is a standard SPI interface.
= Power supply is 4.5V ¨ 5.5V, 3.3V voltage regulator circuit board.
[0087] Microcontroller: The main computing element of the device is an Atmel ATmega128A
microcontroller. The Atmel ATmega128A is a low-power CMOS 8-bit microcontroller based on the AVM) enhanced RISC architecture. By executing powerful instructions in a single clock cycle, the ATmega128A achieves a throughput of nearly 1MIPS per MHz. This allows the system's designer to optimize the device in terms of power consumption and processing speed. Data measurements from sensors and Artificial Intelligence as well as interface control with the operator are the tasks performed by the MCU
in the device. Some of the key features of the device are listed below:
= High-performance, low-power Atmel AVR 8-bit microcontroller.
= Advanced RISC architecture.

Date Recue/Date Received 2022-05-20 = High endurance non-volatile memory segments.
= 128Kbytes of in-system self-programmable Flash program memory.
= 4Kbytes EEPROM.
= 4Kbytes internal SRAM.
[0088] FIG. 5 diagram illustrating a pre-trained MobileNet Deep CNN
classification system. According to FIG. 5, the pre-trained MobileNet Deep Convolutional Neural Network classification system with the help of Tensorflow framework and Keras library were used to classify the type of disease based on the image of psoriasis skin diseases.
[0089] In order to train the network, a suitable database was needed to drive the training process according to the samples. Considering the fact that data collection on medical topics is one of the most challenging issues. The best conditions for images of skin diseases are clinical data collection, but due to lack of time, it has been referred to the reference database. Com.Dermnet website as a complete reference of skin disease images, which includes 23 types of skin diseases, each of which contains training and evaluation data. In the network training process, the training data is used and the accuracy of the network is checked using evaluation data. In this data, there is a group related to psoriasis images which include 2000 color photos regarding different types of psoriasis. According to our requirements, 5 main groups were needed. Studies were conducted on psoriasis and its types, and based on the label of images, the images were manually divided into the 5 groups such as plaque, guttate, pustular, inverse, and erythrodermic for training and evaluation data. A total of 749 samples were obtained for training and 157 samples for evaluation.
[0090] To train the network, due to the small amount of data, the augmentation technique was used, which includes normalization, 90-degree rotation, horizontal and vertical shifts, etc., to strengthen the data. In the architecture used in this project, images with fixed dimensions are required for the input because the input images are of different sizes. First, all images were resized to 224 x 224 dimensions before training. Then input data in these dimensions were examined.
[0091] Due to the fact that the project is related to skin, it is also possible to use (Net Adversarial Generative work) by applying a variety of skin colors to increase and improve the use of data, which can be one of the future works.
[0092] Deep convolutional networks were used for training, which gives good results in image classification. The types of convolutional networks with a different number of convolutional layers were studied to extract the required features using Tensorflow and Keras library.
Pooling layers were used to Date Recue/Date Received 2022-05-20 reduce the increasing dimensions of convolutional. Dense layers were used for classification, and dropout was used to solve overfit problems. The next step was to examine some examples of pre-training networks such as mobilet, Resnet, Vgg, and Alexnet. Resnet50 showed the best performance. Better results were obtained by applying changes in the last layers. However, this model was discarded due to the high memory requirements for this architecture and the weights obtained. MobileNet was used due to the practicality of this project.
[0093] According to this disclosure, the existing weights of the network were not taken into account and retraining was performed. The last 5 layers were eliminated. 6 layers were implemented, including the Flatten, Dense, Dropout, Dense, Dropout, and Denset layers, which turned fine the model for the data project.
[0094] FIG. 6 is a diagram illustrating 86% Accuracy in Classifying Psoriasis Using the MobileNet Deep Neural Network and Augmentation. According to FIG. 6, network training using the optimization function RMSprop with a learning rate of 0001.0 and loss = categorical-crossentropy, performed with 140 IPAC, achieved a classification accuracy of 86% was used. Furthermore, deep neural network MobileNet and the augmentation, 5 types of psoriasis classified with 87% accuracy was also used.
[0095] FIG. 7 is a diagram illustrating a Confusion Matrix. According to FIG.
7, the accuracy of the model is indicated, which after 140 IPAC, the model achieved 95% accuracy with the education data and is shown in blue. For the evaluation data, which the model did not see any of this data during the training, with 86%
was able to recognize and classify the data and is shown shaded in the chart.
[0096] FIG. 7 also evaluates the efficiency of the model. The evaluation data is given to the model one by one and for each, one class is received as output. The predicted class and the actual data class can be displayed in a table.
[0097] FIG. 8A is a table illustrating various evaluation criteria for model testing. For example, out of 23 inverse data samples, 20 of them were found to be inverse, which is correct. 2 samples were found to be Plaque and one of them guttate. If the goal is to test the model for each of the classes, the table in FIG. 8A
is used, which contains various evaluation criteria.
[0098] To describe this better, consider the following case. FIG. 8B is a table illustrating two related outputs of each sample class. According to the table in FIG. 8B, a model that has two outputs, one positive and one negative. These two outputs actually represent the class of each sample.

Date Recue/Date Received 2022-05-20 [0099] When a model predicts a positive result, it wants to know how accurate that result is. If the value of false positives is high, accuracy is a good measure. For example, here the 84% accuracy for Plaque means that the model diagnosed the disease 84% correctly. If the value of false negatives is high, the recall criterion is a good one. The fl criterion is a good criterion for evaluating the accuracy of the model, which is "one"
in the best case and "zero" in the worst case.
[00100] This UV device has two applications. One for the client and the other for the doctor. FIG. 9 is a diagram illustrating Client Application overview. FIG. 10 is a diagram illustrating Doctor Application overview. According to FIG. 9 and FIG. 10, the UV applications are used to treat skin conditions with a UV device. The UV application works as follows.
[00101] After installing the application, the splash page will appear. After a few seconds, the login page will appear. To enter the application, the user should enter the username and password and then click on the login button. If the account is correct, the main page of the application will be accessed. The main page contains the following menus. The menus include the Treatment and Profile pages.
[00102] The treatment page can display the treatment records that the user has already created. By clicking on each record, one can see the details of each record and continue the treatment process. When the user wants to treat an area of the body, s/he creates a treatment record. Then the user clicks the + button and fills out the form that appears. The items in the medical record form include the following: Record name, Body area, Skin color, Doctor name, Underlying diseases, Disease name, and Area size.
[00103] The disease name is recognized by the artificial intelligence network.
To diagnose the disease name, click on the "Add Disease" button and the user will first be asked for the type of treatment. The treatment options include corona and psoriasis options. After selecting the treatment type, if the user is connected to the device camera, the camera image will be displayed, and clicking the Al detection button will diagnose the skin disease.
[00104] After a medical record is created, the record is sent to the doctor of the user's choice and awaits the doctor's approval. When the doctor reviews the record, the number of treatment sessions and treatment days required are determined.
[00105] To start the treatment, the user clicks on the created record then clicks on the active treatment session. Before starting the treatment session, one should select the skin color of the area where the previous session was irradiated. After that, the user clicks the "Calculate Dosage"
button. At this moment, the phone is connected to the device. It calculates the required dose based on the user information and displays it on Date Recue/Date Received 2022-05-20 the screen. Then, clicking on the "Start" button, the command to start the treatment process is sent to the device. The device begins to work.
[00106] After the end of the treatment process, the user clicks on the treatment session and records the end of it. At the beginning of the treatment, if the user selects the skin color of the previous session red, the record will be sent to the doctor. Then the session will need the doctor's approval to start the treatment process.
[00107] On the record page of the list of treatment sessions along with the percentage of progress, the start date of treatment can also be seen. The profile page contains the user information and logout.
[00108] The diagnosis is created by the deep learning artificial intelligence network in the Python language and the Tensorflow Lite library. The way the network works is using the IP
camera. The skin environment is photographed. The image frame is resized in the dimensions of 256 * 256 and stored on the internal memory of the phone, then the network is loaded in the app. After that is sent to the network. The name of the disease is visible in the network output.
[00109] The UV device is used to treat the disease. In the application, when creating a treatment record, the necessary information for using the device is taken from the user. When the treatment starts, this information is sent to the device via Bluetooth connection and the USART
protocol. After receiving the disease information, the device calculates the radiation dose and its working time, then it starts irradiating the radiation on the skin.
[00110] Medical records and treatment sessions are stored on the server using web services. The doctor in charge of the record can monitor the patient's recovery. The user (patient) creates a medical record through the patient application and the record information is sent to the doctor application and the physician can visit the patient record.
[00111] To use the application for doctors, first the login page is opened, the doctor enters his user information and gets to the main page of the application. On the main page there are two tabs, the "Folders"
tab and the "Treatments" tab.
[00112] The Folder tab displays the files referred to the doctor, and the doctor can view the details of each file by selecting it and then confirming the file by entering the treatment start date, number of sessions, and time between sessions.

Date Recue/Date Received 2022-05-20 [00113] The Treatment tab can display the treatment sessions of all patients that need to be confirmed by the doctor. The doctor can confirm the session after seeing the details of the session.
[00114]Records are created by the user application and then stored on the server database using the rest API. The files are visible in the doctor application.
Treatment Dosage:
[00115] FIG. 8C is a table illustrating appropriate dose for each skin color to treat psoriasis. FIG. 8D is a table illustrating appropriate dose to use in corona. According to FIG. 8C and FIG. 8D, the parameters affecting the dose are as follows:
[00116] Treatment: For treatment of psoriasis or coronavirus.
[00117] Patient's skin color: It is divided into 6 general categories from white (pale) to black. The appropriate dose for each skin color for the treatment of psoriasis is shown in tables in FIG. 8C and FIG.
8D.
[00118] Body Area: the choice of body area is important because the maximum dose for the face in any treatment model (corona or psoriasis) is equal to 1000 mJ/cm^2. In other words, the above tables are not correct for the face.
[00119]Number of Treatment Sessions: Such that the patient requires a total of 40 treatment sessions.
Therefore, the dose used in each session varies from the minimum dose to the maximum dose allowed in relation to the number of sessions.
Dosage = (maxdose-mindose) * (current session / total sessions) + mindose.
[00120] For example, if the treatment session is 20 and the minimum and maximum doses are 0.5 and 1 watt/cm, respectively, the dose is calculated as follows:
Dosage = (1-0.5) * (20/40) + 0.5 = 0.75 [00121] Correction Rate for Delayed Sessions: If for any reason the patient has been unable to continue treatment continuously, the dose should be adjusted as follows:
1-week treatment interruption: the last dose should be repeated. In other words, the pre-session will be repeated.

Date Recue/Date Received 2022-05-20 1 to 2 weeks: The previous dose should be reduced by 25%.
2 to 4 weeks: The previous dose should be reduced by 50%.
More than 4 weeks: Treatment should be restarted from the beginning.
[00122] Skin Inflammation: If the skin becomes inflamed and irritated under radiation treatment, the dose of the previous session should be applied without an increase. If the inflammation is very severe, the application will automatically notify the doctor and the treatment will be stopped.
[00123] In general, taking into account all the above parameters, the formula for calculating the dose in the device will be as follows:
if (region ¨ 0)! * face * /
maxdose = 1.00;
else / * other body regions * /
maxdose = DOSEMAX [curetype] [skincolor];
Dosage = (((maxdose - DOSEMIN [curetype] [skincolor]) * (float) (session-erythema- (correction>
0)) / 39.0) + DOSEMIN [curetype] [skincolor]) * ((100.0- (float) correction) /100.0);
[00124] The maximum and minimum values are defined as follows (in J/cm^2):
float DOSEMAX [2] [6] = {0.70,1.700,2.500,2.500,3.400,3.4001, / * corona * /
{2.000,2.000,3.000,3.000,5.000,5.000}1; / * for psoriasis * /
float DOSEMIN [2] [6] = {{.300, .300, .500, .500, .800, .800}, / * corona * /
{.300, .300, .500, .500, .800, .800} l; / * for psoriasis * /
[00125] UV Bluetooth Connection: The device transmits the following string once every two seconds:
# UV **
After selecting the automatic mode, the parameters for color, area, etc.
should be sent from the phone as follows:
#cmd cure = 0 skin = 3 regi = 1 dise = 0 sess = 1 con = 0 eryt = 0 size = 0 **
Where the letters cmd indicate the command sent to the device and cure: Curetype: corona treatment 0 or psoriasis 1 Date Recue/Date Received 2022-05-20 skin: skincolor: skin color (number between 0 and 5) regi: region: (number between 0 and 4) body area dise: disease: (number between 0 and 5) type of disease sess: session: multiple sessions (number between 1 and 40) con: correction: correction factor for forgotten sessions (number between 0 and 100) erythema: erythema (number between 0 and 2) size: areasize: has no effect on the formula [00126] Once the above parameters are received, the dose is calculated and in addition to the daily display LCD, the following device string is sent to the phone containing the calculated dose to be displayed to the user of the application.
# dosage = 1.61 **
The dose number is a decimal number with two decimal places.
The device then begins to operate by receiving the following confirmation string from the phone:
# confirm **
[00127] Implementations disclosed herein provide systems, methods and apparatus for generating or augmenting training data sets for machine learning training. The functions described herein may be stored as one or more instructions on a processor-readable or computer-readable medium. The term "computer-readable medium" refers to any available medium that can be accessed by a computer or processor. By way of example, and not limitation, such a medium may comprise RAM, ROM, EEPROM, flash memory, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer. It should be noted that a computer-readable medium may be tangible and non-transitory. As used herein, the term "code" may refer to software, instructions, code or data that is/are executable by a computing device or processor. A "module" can be considered as a processor executing computer-readable code.
[00128] A processor as described herein can be a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A
general purpose processor can Date Recue/Date Received 2022-05-20 be a microprocessor, but in the alternative, the processor can be a controller, or microcontroller, combinations of the same, or the like. A processor can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor may also include primarily analog components. For example, any of the signal processing algorithms described herein may be implemented in analog circuitry. In some embodiments, a processor can be a graphics processing unit (GPU). The parallel processing capabilities of GPUs can reduce the amount of time for training and using neural networks (and other machine learning models) compared to central processing units (CPUs). In some embodiments, a processor can be an ASIC including dedicated machine learning circuitry custom-build for one or both of model training and model inference.
[00129] The disclosed or illustrated tasks can be distributed across multiple processors or computing devices of a computer system, including computing devices that are geographically distributed. The methods disclosed herein comprise one or more steps or actions for achieving the described method. The method steps and/or actions may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is required for proper operation of the method that is being described, the order and/or use of specific steps and/or actions may be modified without departing from the scope of the claims.
[00130] As used herein, the term "plurality" denotes two or more. For example, a plurality of components indicates two or more components. The term "determining" encompasses a wide variety of actions and, therefore, "determining" can include calculating, computing, processing, deriving, investigating, looking up (e.g., looking up in a table, a database or another data structure), ascertaining and the like. Also, "determining" can include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory) and the like.
[00131] The phrase "based on" does not mean "based only on," unless expressly specified otherwise. In other words, the phrase "based on" describes both "based only on" and "based at least on." While the foregoing written description of the system enables one of ordinary skill to make and use what is considered presently to be the best mode thereof, those of ordinary skill will understand and appreciate the existence of variations, combinations, and equivalents of the specific embodiment, method, and examples herein. The system should therefore not be limited by the above described embodiment, method, and examples, but by all embodiments and methods within the scope and spirit of the system. Thus, the present disclosure is not intended to be limited to the implementations shown herein but is to be accorded the widest scope consistent Date Recue/Date Received 2022-05-20 with the principles and novel features disclosed herein.

Date Recue/Date Received 2022-05-20

Claims (18)

Claims We Claim:

1. A portable phototherapy device for the treatment of psoriasis, the device comprising:
a housing;
a processor;
a circuit board;
a display;
a UV lamp configured to support narrow-band ultraviolet-B (UVB);
a ballast configured to control the current to the lamp;
a camera for capturing and processing images;
a distance sensor;
a Bluetooth module configured to enable the device to communicate over Bluetooth ; and a smartphone adapter configured to connect to a smartphone placed on the housing;
wherein the device is configured to wirelessly communicate to the smartphone via Bluetooth and synchronize with a smartphone application with data captured from the camera.

2. The device of Claim 1 further comprising a power switch, AC socket and fan.

3. A method of capturing images with a portable phototherapy device comprising the steps of:
capturing an input image of a person's skin with a camera of the device;
processing the input image using artificial intelligence (AI) and an IoT
program on the device;
sending the data wirelessly to a smartphone using Bluetooth ; and displaying the data on the smartphone display.

4. The method of Claim 3 further comprising the step of training a neural network to strengthen the data using augmentation techniques selected from a list consisting of normalization, 90-degree rotation, horizontal and vertical shifts.

5. The method of Claim 4 further comprising the step of resizing the input images to a fixed dimension of 224 x 224 dimension prior to training.

Date Recue/Date Received 2022-05-20

6. The method of Claim 4 further comprising using deep convolutional networks for training, which gives good results in image classification.

7. The method of Claim 6 further comprising using Tensorflow and Keras library with a different number of convolutional layers to extract the required features.

8. The method of Claim 6 further comprising using pooling layers to reduce the increasing dimensions of convolutional whereby dense layers were used for classification and dropout were used to solve overfit problems.

9. The method of Claim 3 further comprising using pre-training networks such as Mobilet, Resnet, Vgg, and Alexnet.

10. The method of Claim 9 further comprising adding new layers to the pre-trained network to improve the performance of the network.

11. The method of Claim 9 wherein the existing weights of the network were not taken into account and retraining was performed whereby the last 5 layers were eliminated and 6 layers were implemented, including the Flatten, Dense, Dropout, Dense, Dropout, and Denset layers, which turned fine the model for the data project.
12. The method of Claim 3 wherein network training using the optimization function RIVISprop is used with a learning rate of 0001.0 and loss = categorical-crossentropy, performed with 140 IPAC, achieving a classification accuracy of 86%.
13. The method of Claim 12 wherein after 140 IPAC, the model achieved 95%
accuracy with the trained data and for the evaluation data, which model haven't seen any of this data during the training, with 86% were able to recognize and classify the data.
14. The method of Claim 3 wherein the processing step further comprises using a deep learning artificial intelligence network in the Python language and the TensorFlowLite library.

12. The method of Claim 14 further comprising resizing the image to an image frame of dimensions of 256 by 256 and saving the image in the internal memory of the smartphone.

13. The method of Claim 3 further comprising loading the neural network on an application on the smartphone and displaying the name of the disease on the smartphone.
Date Recue/Date Received 2022-05-20

14. The method of Claim 13 further comprising creating a treatment record, whereby necessary information for using the device is taken from the user.

15. The method of Claim 3 further comprising sending information from the device to the smartphone via a Bluetooth connection and the USART protocol.

16. The method of Claim 14 wherein after receiving the disease information, the device calculates the radiation dose and its working time and provides this date to the user.

17. The method of Claim 3 wherein medical records and treatment sessions are stored on the server using web services whereby the doctor in charge of the record can monitor the patient's recovery.

18. The method of Claim 17 wherein the medical records are created by the user application and then stored on the server database using the rest API whereby the files are visible in the doctor's application.

Date Recue/Date Received 2022-05-20