AU2021103658A4

AU2021103658A4 - Iot based smart pandemic detection device and method for pandemic preventive measures

Info

Publication number: AU2021103658A4
Application number: AU2021103658A
Authority: AU
Inventors: Himadri Biswas; Arup Roy; Manash Sarkar; Vaskar SARKAR; Govind Verma
Original assignee: Sarkar Manash Mr; Sarkar Vaskar Mr; Verma Govind Mr
Current assignee: Sarkar Manash Mr; Sarkar Vaskar Mr; Verma Govind Mr
Priority date: 2021-06-27
Filing date: 2021-06-27
Publication date: 2022-03-31
Anticipated expiration: 2029-06-27

Abstract

The present invention provides for an IoT-based smart pandemic detection device and method based on a cloud-oriented architecture - Pandemic Preventive Measure as a Service (PPMaaS) network for early tracking of any pandemic disease affected populations, deploying in crowd transition areas and taking precautions by contacting nearest relevant hospitals to access treatment immediately based on the condition of the patients, wherein the device based on an Al enabled cloud computing based framework comprising Smart Pandemic Determiner (SPD) under Pandemic Service Controller (PSC) expertise, operates from sample collection to detection phase based on the available resources provided by the respective cloud service provide (CSP) and sends the result to the Government Health Administration System (GHAS) through nearest Sanatorium or Health Centre (SHC) for preventive measure. Page 4 of 4 START No PSC checks for SSD's authentication & SLA Yes CSP checks for PSC's No authentication & SL A Analysed Result<= Threshold No Yes Yes Check the availability of the Client 4No Figure 4. Flowchart of PPMaaS Model

Description

Page 4 of 4

START

No PSC checks for SSD's authentication & SLA

Yes

CSP checks for PSC's No authentication & SL A

Analysed Result<= Threshold No Yes

Yes Check the availability of the Client

4No

Figure 4. Flowchart of PPMaaS Model

IOT BASED SMART PANDEMIC DETECTION DEVICE AND METHOD FOR PANDEMIC PREVENTIVE MEASURES FIELD OF THE INVENTION

The present invention provides for an IoT-based smart pandemic detection device and method based on a cloud-oriented architecture - Pandemic Preventive Measure as a Service (PPMaaS) network for early tracking of any pandemic disease affected populations, deploying in crowd transition areas and taking precautions by contacting nearest relevant hospitals to access treatment immediately based on the condition of the patients, wherein the device based on an Al enabled cloud computing based framework comprising Smart Pandemic Determiner (SPD) under Pandemic Service Controller (PSC) expertise, operates from sample collection to detection phase based on the available resources provided by the respective cloud service provide (CSP) and sends the result to the Government Health Administration System (GHAS) through nearest Sanatorium or Health Centre (SHC) for preventive measure.

BACKGROUND ART

Through a historical viewpoint, the world has seen a number of devastating occurrences at various times, some with unparalleled impacts. One of the worst cases was the Black Death of the 14th century that killed an estimated 100-200 million people (Duncan C J, Scott S 2005 What caused the black death? Postgrad. Med. J., 81, 315. (CrossRef) (PubMed)). Primary influenza outbreaks (Flu) have occurred in recent years, including: the first (HINI influenza virus, or 'Spanish Flu') reported in 1918 (around 500 million deaths), originating in Etaples, France (Martini M, Gazzaniga V, Bragazzi N L, BarberisI2019 The spanish influenza pandemic: A lesson from history 100 years after 1918 J. Prev. Med. Hyg., 60, E64-E67); followed by influenza A subtype H2N2 ('Asian Flu') originating in China in 1957-1958 (1.1 million deaths); followed by influenza A (H3N2) originating in China in 1968 (1 million deaths) Then, in the United States, the 2009 HINI influenza (or swine flu) virus (12,469 deaths). Other outbreaks include SARS (or Severe Acute Respiratory Syndrome), which broke out in 2003 (774 deaths) in China's Guangdong Province; the 2014 Ebola (Zaire ebola virus) virus (11,315) in Zaire (now the Democratic Republic of Congo); the 2015 Zika Virus spread by the Brazilian Aedes aegypti mosquito; and the latest corona virus in 2020 that originated in Wuhan, China.

The whole world has focused closely on an outbreak of respiratory illness caused by a novel corona virus. The virus that causes COVID-19 is transmitted mainly by droplets that are generated when a person is infected with cough, sneezing, or exhales.

The viral disease COVID-19 (Corona virus disease 2019) with the symptom of extreme acute respiratory syndrome corona virus 2 (SARS-CoV-2) was first detected at Wuhan, China in December 2019, resulting in a continuing pandemic. There were millions cases reported since June 2020, resulting in huge deaths across 188 countries. There are no full-proof vaccines or effective antiviral therapies for COVID-19 according to the World Health Organization (WHO) report, though some vaccines are available.

In this pandemic scenario, a compulsory test must be considered for a large number of citizens in all countries, which may not be accessible for all individuals or may not be supported by the government due to the availability and expense of the test kit or any other testing tool or procedures that require time to install and use it for community use.

Therefore, there is a need in the art for an improved IoT-based smart pandemic detection device and method for pandemic prevention and control method.

SUMMARY OF THE INVENTION

In an aspect according to the present invention, it relates to an IoT-based smart pandemic detection device on a cloud-oriented architecture - Pandemic Preventive Measure as a Service (PPMaaS) network for early tracking of any pandemic disease affected populations, deploying in crowd transition areas and taking precautions by contacting nearest relevant hospitals to access treatment immediately based on the condition of the patients, wherein the device based on an Al enabled cloud computing based framework comprising Smart Pandemic Determiner (SPD) under Pandemic Service Controller (PSC) expertise, operates from sample collection to detection phase based on the available resources provided by the respective cloud service provide (CSP) and sends the result to the Government Health Administration System (GHAS) through nearest Sanatorium or Health Centre (SHC) for preventive measure.

In an embodiment, the device collects data from wide range of sensors such as temperature, cameras, inertial, proximity, microphones, colour, humidity, and likes.

In an embodiment, the Al enabled cloud computing based framework combines the usage of the sensors with cloud environment, converted it into a smarter portable system to gain more access and analyse power for COVID-19 like pandemic detection.

In an embodiment of the device according to the present invention, - to calculate different health issues, such as respiratory rate, heart rate, heart rate variability and health issues like skin & eye disease a advanced day-to-day sensors is used; - to track both the 'heart rate' and the 'rate of Respiration' the front and rear camera sensors of a smart phone is used and wherein 'photoplethysmogram (PPG) fingertip signal' from the top of the 'rear-facing camera' is used to achieve the heart rate; - for estimation of heart rate and heart rate variability through bare skin visual 'photoplethysmogram (PPG) signal' camera and microphone are mutually being used; - to predict fever, the sensor temperature-fingerprint is used; - to detect nasal symptoms, the application software Listen-to-Nose is used; - to test the lung function "SpiroSmart" is microphone based application software is used; - to detect a coughing and respiratory sound microphone sensor is used; - to assess mental health and physical activity and movement, embedded sensors like accelerometer, light, GPS, microphone etc. are used.

In an embodiment, the Cloud Service Provider (CSP) provides storage or software or hardware or network infrastructure to the customers through SaaS or IaaS or PaaS.

In an embodiment, the Government can create a special unit to track and deal with the pandemic situation.

In an embodiment, for spot diagnosis of the disease, a small device made by sensors connected to the cloud environment is used to sense data.

In an embodiment, for maintaining a smooth contact between SSD and CSP, PSC plays a major role as it compelled to register SSD and CSP's before assigning resources to the cloud server and to avoid potential conflict; and wherein after receiving the periodic signals from the CSPs, SPD will use the available cloud resources and will be able to collect data from SSD, analyze it and eventually identify the disease itself.

In an embodiment, the device provides for a portable testing device with an efficient interface that can be incorporated as part of the patient safety management system into medical and home equipment

In another aspect, the invention provide for an IoT-based smart pandemic detection method on a cloud-oriented architecture - Pandemic Preventive Measure as a Service (PPMaaS) network for early tracking of any pandemic disease affected populations, deploying in crowd transition areas and taking precautions by contacting nearest relevant hospitals to access treatment immediately based on the condition of the patients, wherein it comprises the following SCPD (Smart Cloud Pandemic detection) algorithm: 1. Pandemic Service Controller (PSC) offers preliminary detection of Pandemic disease using Smart Screening Device (SSD) through Smart Pandemic Determiner (SPD) in cloud environment 2. SSD may be installed in important places 3. SSD collects symptoms from the Clients 4. SSD requests to PSC for diagnosing 5. PSC checks for the SSD's authentication and Service Level Agreements (SLA) 6. If Authentic 6.1 Service accepted and acknowledgement sends to the SSD 6.2 PSC requests to the intended CSP for getting the on-demand cloud resources 6.3 CSP checks for the PSC's authentication and SLA 6.4 If bona fide 6.4.1 Service established and acceptance message sends to PSC 6.4.2 Based on the agreement CSP sends the periodic signals / resource details to Smart Pandemic Determiner (SPD) 6.4.3 SPD store the record details in a log file 6.4.4 Go to step 8 6.5 Else 6.5.1 Message - "Not acknowledged for services" back to the PSC 6.5.2 Go to step 6.2 7. Else 7.1 Message - "Not accepted for services" returns to the SSD 7.2 Go to step 4

8. SPD collects the client's symptoms from the SSD by its Fetching module 9. SPD fetches the periodic record details about the resources from its log file 10. Based on the collected symptoms SPD analysis on the data by its Testing module 10.1 If within Threshold 10.1.1 "Not Infected" -- message returned to the SSD as well as to the Client 10.1.2 Go to step 11 10.2 Else 10.2.1 SPD trace the current location of the infected person 10.2.2 "Infected" -- message returned to the SSD, to the Client, as well as to the nearest Sanatorium / Health Centre (SHC) with details about the Client 11. Check for the availability of the client 12. If client exists 12.1 Go to step 3 13. Else 13.1 Go to step 14 14. SHC sends the pandemic reports to the Government Health Management System (GHAS) 15. GHAS takes necessary actions 16. End

DETAILED DESCRIPTION OF THE DRAWINGS

Figure 1. represents the schematic representation of the proposed Service Model (PPMaaS) including basic Cloud Service Model

Figure 2. represents the schematic representation of the proposed Architecture of Pandemic Preventive Measure as a service (PPMaaS) Model

Figure 3. represents the Process Flow Diagram of Smart Pandemic Determiner (SPD)

Figure 4. represents the Flowchart of PPMaaS Model

DETAILED DESCRIPTION

The present invention is directed towards an IoT-based smart pandemic detection device on a cloud-oriented architecture - Pandemic Preventive Measure as a Service (PPMaaS) network for early tracking of any pandemic disease (such as COVID19) affected populations, deploying in crowd transition areas and taking precautions by contacting nearest relevant hospitals to access treatment immediately based on the condition of the patients.

For this proposed IoT-based Smart Pandemic Detection Device deploying low-cost and less effort in various heavy public transition areas, an infected person can be monitored remotely from anywhere in the world and appropriate steps can be taken to apply health / medical recommendations recommended by competent authorities to spotted infected individuals. Through this smart screening process, we will protect from cross-infection other uninfected persons. Therefore, the government may take the steps for compulsory installation of such IOT based smart device as a preventive action at appropriate places to identify and track the infected persons.

The world has seen a variety of catastrophic events at various times from a historical point of view, some with unprecedented impacts, as discussed above. The novel corona virus (COVID 19) of respiratory disease - an outbreak was closely targeted worldwide which is transmitted mainly through droplets formed when a person is infected with cough, sneezing, or exhales. One can get infected with breathing in the virus if he or she is close to someone affected by COVID-19 or touches a contaminated surface by eyes, nose or mouth. The common symptoms including cough, Fever, shortness of breath, fatigue, smell, taste and loss of sense, but some patients often suffer from anxiety and discomfort, inflammation of the nose, runny nose, sore throat or diarrhea. Some individuals affected with COVID-19 have mild symptoms breathing problems, and they live without medication but elderly people and those with ongoing health issues such as heart disease, diabetes, chronic disease and cancer are also more likely to develop serious medical condition. Within the first three days of symptom onset, it is most contagious but it can widen even from people who have no symptoms before symptoms start (How COVID-19 Spreads 2 April 2020 Centers for Disease Control and Prevention (CDC) (Archivedfrom the original on 3 April 2020. Retrieved 3 April 2020) COVID 19 length of the incubation (time from symptoms to infection) varies between 1 and 14 days. According to WHO, almost 80 per cent of confirmed patients recover from the disease without significant complications, but one in six users of COVID-19 will get critically ill and encounter breathing difficulties. In more serious cases, disease can lead to acute pneumonia and other complications which can be handled only in higher-degree facilities (District Hospitals and above). In a few cases, this can also cause death.

According to an embodiment, the device and the method according to the present invention would prevent and slow down transmission of COVID-19 or similar contagious diseases for mass people on the spot by smart screening devices that can help diagnose this disease early on. A wide range of sensors such as temperature, cameras, inertial, proximity, microphones, colour, humidity, and many others are available on the market. The inventors of the present invention have combined the usage of these sensors with cloud environment, converted it into a smarter portable system to gain more access and analyse power for COVID-19 like pandemic detection.

In the Al enabled Cloud computing based proposed framework used in the device and the method according to the present invention, the Smart Pandemic Determiner (SPD) under Pandemic Service Controller (PSC) expertise not only from sample collection to detection phase based on the available resources provided by the respective Cloud Service Provider (CSP), and finally take the responsibility to send the result to the Government Health Administration System (GHAS) through nearest Sanatorium or Health Centre (SHC) for preventive measure.

People fight the pandemics at different times from ancient times to dates. Recently, the entire world has concentrated closely on an epidemic of respiratory disease triggered by a novel corona virus (COVID-19), get troubled worldwide, and kill the lives of people .Governments in a number of countries have taken measures to reduce the COVID-19 pandemic effect. During this immense and chaotic time, globally, scientists are working relentlessly to find a vaccine. Candidates for the COVID-19 vaccine, as well as other candidates for preclinical production and testing are currently in step 1-3 studies. In this circumstance almost all countries around the world are struggling together just to monitor the corona virus outbreak COVID-19. So many techniques such as "NAT (Nucleic Acid Test)", and "CT (Computed Tomography)" are available to diagnose COVID-19 corona virus disease, where NAT is used to classify various sequences of nucleic acid and organism types, specifically blood, tissue, or urine related diseases caused by a virus or bacteria. On the other hand, CT scanning is the most efficient and realistic way to detect the extent and degree of inflammation of the lung (Wang D, Hu B, Hu C, Zhu F, Liu X, Zhang J, Wang B, Xiang H, Cheng Z, Xiong Y et al. 2020 (Clinical characteristicsof 138 hospitalizedpatients with 2019 novel coronavirus-infectedpneumonia in wuhan, china)), whereas detection kits and NAT technique are becoming critical to detect COVID-19 virus. The inclusion of regular clinical diagnostic presentations for radiographic pneumonia in the province of Hubei reported by National Health Commission of China (http://www.chinadaily.com.cn/m/chinahealth/index.html) as to diagnosis the severity of COVID-19 pneumonia by the key images of CT scan. The pandemic COVID-19 and the resulting enormous demand for treatment have inspired companies, academics and researchers to develop methods of detection that are highly efficient, smarter, and more reliable. A smart method of CT image reading system for the COVID-19 revealed by Ping, a Smart Healthcare device that can interpret findings at a precision rate above 90 percent in around 15 seconds (https:/www.bioworld.com/articles/433530-china-uses-ai-in-medical-imaging-to-speed up-covid-19-diagnosis). On the other side, of course neither the 'Reverse Transcription Polymerase Chain Reaction (RT-PCR)' nor the COVID-19 'CT scans' diagnosis are well suited (Bini S A 2018 Artificial intelligence, machine learning, deep learning, and cognitive computing: What do these terms mean and how will they impact health care? J Arthroplasty, 33, 2358-2361). In fact, with advances in computational capabilities and widespread use of Al, machine learning, big data like technologies embedded with cloud computing, it has become possible to collect and analyse vast amounts of data from various sources in real time, and to make informative predictions of it (Azmak , Bayer H, Caplin A, Glimcher P, Koonin S, PatrinosA 2015 Using big data to understand the human condition: The kavli human project. Big Data, 3, 173-188). A deep learning based Al algorithm with high resolution CT images (Chen J, Wu L, Zhang J, Zhang L, Gong D, Zhao Y, Hu S, Wang Y, Hu X, Zheng B et al. 2020 Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography: a prospective study medRxiv) has developed for detection of COVID-19. For detection of COVID-19, CT images of the volumetric chest analysed by 'COVNet (COVID-19 neural network detection)', the three dimensional deep learning technique has been used (Li L, Qin L, Xu Z, Yin Y, Wang X, Kong B, Cao K 2020. Artificial intelligence distinguishes COVID-19 from community acquiredpneumonia on chest CT. Radiology, 200905). A convolutionary ResNet-50 model called COVNet which takes its inputs as a sequence of CT segments and the CT image class labels determines its output and demands the proposed model is more capable to detect COVID-19 due to its AUC value 0.96 (He K, Zhang X, Ren S, Sun J 2016 Deep residual learningforimage recognition. In Proc. of the IEEE Conf on Computer Vision and PatternRecognition (pp. 770-778)). To detect cases of corona virus using pulmonary CT images another location-attention system concatenation with the three dimensional CNN ResNet-18 network - a deep learning method is proposed in (Xu X, JiangX, Ma C, Du P, Li X, Lv S, Li Y 2020. Deep learning system to screen coronavirus disease 2019 pneumonia arXivpreprintarXiv:2002.09334). Various types of COVID-19 CT imaging were found in (Kanne J P 2020 Chest CT findings in 2019 novel coronavirus (2019-nCoV) infections from Wuhan, China: key points for the radiologist. Radiology, 200241). Another deep learning model derived from customized stacked auto encoder is used to guess COVID-19 cases throughout China.

For assessing the risk of infection at the community level for a given geographic area, a-Satellite another Al-based prototype system is proposed (Ye Y, Hou S, Fan Y, Qian Y, Zhang Y, Sun S, Laparo K 2020 Satellite: An AI-driven system and benchmark datasetsforhierarchicalcommunity level risk assessment to help combat COVID-19. arXiv preprint arXiv:2003.12232). Social media data for a specified region could be inadequate to being enriched by the conditional adversarial generative networks for knowing COVID-19's public awareness (Mirza M, Osindero S 2014 Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784). To estimate the risk indexes aggregate information from the given city's locality areas, an auto encoder model of a heterogeneous graph is then configured. Again, a stochastic agent-based discrete model called "ACEMod (Australian Census-based Epidemic Model)" is used to model the COVID-19 pandemic, previously used for influenza pandemic simulation, based on main disease transmission parameters across Australia (Cliff 0 M, Harding N, Piraveenan M, ErtenE. Y, Gambhir M, Prokopenko M 2018 Investigatingspatiotemporaldynamics and synchrony of influenza epidemics in Australia:An agent-based modelling approach. Simulation Modelling Practice and Theory, 87, 412-431). The best solution the model suggests is to incorporate foreign entry, case separation and social distancing constraints with 80 percent or higher enforcement in at least 13 weeks. For COVID-19 infection risk prediction, incorporating the concept of susceptible infected disease a hybrid Al model is proposed. For better understanding and manage the worldwide public health during the COVID-19 pandemic, another Al model based upon some standardisation protocols and data sharing techniques is proposed (Allam Z, and Jones D S 2020, March On the coronavirus (COVID 19) outbreak and the smart city network: universal data sharing standards coupled with artificial intelligence (AI) to benefit urban health monitoring and management. In Healthcare (vol. 8, no. 1, p. 46)). In addition the COVID-19 is detected by means of medical detection kits. But that method is expensive, requiring a diagnosis to be installed. Now a day, a lot of sensors on the market with strong computing capabilities can be constantly sensed awareness of day-to-day activities. For instance, the temperature-fingerprint sensor can be used to sense the body temperature to predict fever level (Maddah E, and Beigzadeh B 2020. Use of a smartphone thermometer to monitor thermal conductivity changes in diabeticfoot ulcers: a pilot study. Journal of Wound Care, 29(1), 61-66). On the other hand, inertial sensors can be used to detect human fatigue level where Images and videos taken from the camera or data taken from on board inertial sensors. Similarly, Story et al. (Aldridge R W, Smith C M, Garber E, Hall J, Ferenando G, AbubakarI2019 Smartphone enabled video-observed versus directly observed treatmentfor tuberculosis: a multicentre, analyst blinded, randomised, controlled superiority trial. The Lancet, 393(10177), 1216-1224) use Smartphone videos to predict nausea, while use camera images and measurements of inertial sensors used by Lawanont et al. (Inoue M, Mongkolnam P, Nukoolkit C 2018 Neck posture monitoring system based on image detection and smartphone sensors using the prolonged usage classification concept IEEJ Transactions on Electrical and Electronic Engineering, 13(10), 1501 1510) to monitor neck location and predict headache rates in humans. Audio data obtained from the microphone sensor is also used for detecting the form of cough.

Considering all the scenarios the present inventors have come up with a smart screening system that will allow people to check in different heavy public transition areas to avoid gathering in hospitals or test centres that not only reduce the risk of cross-infection with others, save the cost of test kits, and also identify the places where the infected person travels. The present invention also aims to avoid the spread of the disease as soon as possible and the whole thing will be managed through cloud computing by the Pandemic Service Controller (PSC).

1.1. Sensors: To calculate different health issues, such as respiratory rate, heart rate, heart rate variability and health issues like skin & eye disease advanced day-to-day sensors can be used. Typical health monitoring sensors are shown in Table 1.

Table 1. Typical Health Monitoring Sensors

Typical Sensors Health Issues Microphone Nasal symptoms (Blowing the nose, Sneezing and Runny Nose), Lung Functions, Ear health, chronic pulmonary diseases such as cough, asthma, shortness of breath, Fatigue level Image Sensor (Camera), Microphone Cardiovascular activity - Heart Rate, Heart Rate Variability, Respiratory and Lung Health Image Sensor (Camera) Eye Health, Skin Health Temperature, Thermal Camera Body Temperature Measurement Motion sensors (Accelerometer, Gyroscope, Physical Activity and Movements Proximity), GPS Motion sensors (Accelerometer, Gyroscope), Camera, Cognitive function and Metal health Assessment Light Sensor, GPS Motion sensors (Accelerometer, Gyroscope) Sleep GPS Track Location

Utilization ofsome typical sensors to detect the common symptoms of the diseases, like COVID-19:

3.1.1. Camera sensor: HeartRate (HR) & Rate ofRespiration (RR) assessment

Both front and rear camera sensors of smart phone were used to track both the 'heart rate' and the 'rate of Respiration' [35], where 'photoplethysmogram (PPG) fingertip signal' (from the top of the 'rear-facing camera') used to achieve the HR, In which RR is measured from the front camera by tracking chest and abdominal motions. Another way to obtain HR and RR Welch's power spectral density is required by prevailing intensity in the frequency-domain of images. On the other hand to gain 6-60 breaths (wide dynamic range) per minute as an automated selection protocol for 'Region-of-Interest (ROI)' was used. Above mentioned contact based HR monitoring systems enabling the user to direct contact through the smart phone's camera lens holding the fingertip tightly, otherwise erroneous result will obtained due to any change in finger position. At the other hand, contactless monitoring systems measure HR from the face-video-derived PPG signal. In reference Kwon S, Kim H, Park K S 2012 (Validation ofheart rate extraction using video imaging on a built-in camera system of a smartphone In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, CA, USA, 28 August-i September 2012), such a contactless cardiac pulse monitoring device, called 'Face Beat', was introduced. 'Face Beat' measures the heart rhythms and determines HR from the face video taken by a user using the front camera of the smart phone. Reflected light of the increasing blood flow in the facial blood vessels on a specific section of face is detected by the photo-detector image sensor device to calculate the difference and simultaneously captured Green Channel footage data is securitized. Another proposed method Sanyal S, Nundy K K 2018 (Algorithms for Monitoring Heart Rate and RespiratoryRatefrom the Video of aUser's Face. IEEE J Transl. Eng. Heal. Med. 6, 1-11) that the colour of the reflected light is identified from the face and make the difference to obtain both RR and HR.

3.1.2. Microphone and camera sensor: HR and HR Variability (HRV) assessment

Camera and microphone are mutually being used for estimation of HR and HRV through bare skin visual 'photoplethysmogram (PPG) signal', such as the face or fingertip. The properties of haemoglobin light absorption in the blood vary from those of other body tissues including flesh and bone. PPG measures volumetric fluctuations by detecting refractive variations and/or light reflectivity with arterial pulsation through the tissue; whereas red light sources near-infrared (NIR) are used in most commercial devices. Some researchers used an embedded smart phone with a white flashlight to illuminate the PPG tissue. In reference Koenig N, Seeck A, Eckstein J, Mainka A, Huebner T, Voss A, Weber S 2016 (Validation of a New HeartRate Measurement Algorithmfor

Fingertip Recording of Video Signals with Smartphones. Telemed. e-Health, 22, 631-636) a flashlight and camera sensor from the PPG signal obtained from the fingertips were used to test the HR and HRV. The pulsing signal was extracted from the green channel of the video data after a low-frequency band-pass filtering. In corrupt video data commonly recognized by movable objects is identified and discarded. The PPG signal was extracted from all three channels (red, blue and gray) from the index fingertip video, make the difference of maximum and minimal intensity to derive the threshold then based on the threshold value aggregates the pixel values with higher intensity than the threshold. The PPG signal achieved via the red channel other than two channels the regular increase in blood flow rates during the cardiac cycles was reflected and finally applying 'Fast Fourier Transform (FFT)' analysis on this PPG signal the pulse rate was obtained at an accuracy rate about 98 percent.

3.1.3. Temperature or Thermal Camera sensor: Measure the Body Temperature

The sensor temperature-fingerprint, used to predict fever rates. Additional 'ATmega328' microcontroller for tracking patient body temperature (LM35 Temperature Sensor) and pulse (LM358 Heart Beat Sensor) is used as a CPU. In Asaduzzaman M, Hussain K, Tanveer S R et al, December 2015 (Continuous heart rate and body temperature monitoring system using Arduino UNO andAndroid device, Conference Paper - DOI: 10.1109/EICT.2015.7391943) the definition of an embedded system is demonstrated for calculating heart rate and body temperature. Another temperature monitoring sensor the thermal camera embedded with a wristband calculates the reference point temperature. In Tag B, Chernyshov G, Kunze K, 2017 (FacialTemperature Sensing on Smart Eyewearfor Affective Computing, UBICOMP/ISWC '17ADJUNCT, SEPTEMBER 11-15, 2017, MAUI, HAWAII, USA, https://doi.org/10.1145/3123024.3123084) an eye glass tracking system is implemented to follow optimistic cognitive and impulsive conditions for sentimental computing using 'Thermal IR Camera' (contactless temperature sensors) which follows changes in facial temperature.

3.1.4. Microphone Sensor: Detect Nasal symptoms (Blowing the nose, Sneezing andRunny Nose)

The application software Listen-to-Nose detects nasal symptoms that decisive on audio. The application detects sound patterns through a microphone sensor by using acoustic recognition algorithm that match blowing or sneezing (discards other audio information like speech and silence), to identify the symptoms the user has.

3.1.5. Microphone Sensor: Measure Lung Function using

"SpiroSmart" is microphone based application software that sends the recorded audio data of exhalation to a server for calculating the flow rate of exhalation to test the lung function (Chen N C, Wang K C, Chu H H, 2012 Listen-to-Nose: A low-cost system to record nasal symptoms in daily life, UbiComp'12, September 5-8, 2012, Pittsburgh, USA. ACM 978-1-4503-1224-0/12/09, https://doi.org/10.1145/2370216.2370319).A typical spirometer calculates the air flow rate when going through a mouthpiece. Such mouthpiece flow can be combined to achieve Flow vs. Time (FT), Volume vs. Time (VT), or Flow vs. Volume (FV) expiry plots. Different amounts are determined from the plot :(1) 'FVC (Forced Vital Capacity)' - During the expiry period entire volume expelled; (2) 'FEVI (Forced Expiratory Volume)' - In the first second the amount expired; (3) The ratio of FVC & and FEVI (4) 'PEF (Peak Expiratory Flow)' - The best flow rate obtained for the period of the test. Different lung dysfunctions like Mild (60-79%), Moderate (40-59%) and Severe (<40%) are obtained from the combination of flow vs. volume curves.

3.1.6. Microphone Sensor: Detect chronic respiratoryconditions (like cough, asthma, shortness of breath), pulmonary disruption and lung cancer

For faster pulmonary health assessment many researchers used a microphone sensor to detect a coughing and respiratory sound, and analysed the captured audio signals. Using built-in microphone of a smart phone based interactive game 'Flappy Breath' was developed where the users play by inhaling and exhaling to detect breathing. The game measures the occurrence and intensity of the sound inputted when playing using the microphone, determines the typical amount of air flowing referring to silence, inhalation and exhalation. To detect the coughing situation, the audio signal recorded by the smart phone's microphone was analyzed using a cough detection algorithm 'PCA' (Principal component analysis), where the cough sound spectrogram was separated, normalised and analysed on the basis of the distinct pattern and after re-generate the signal of coughing identify the coughing occurrences finally.

3.1.7. Accelerometer, gyroscopes, magnetometers, GPS, Light & Microphone Sensors: Assessment ofMental Health and PhysicalActivity & Movement

Due to variety of inbuilt embedded sensors (like accelerometer, light, GPS, microphone etc.) of today's smart phones, data such as call history, SMS history and system utilisation to track a person's behaviour remotely and determine their mental health or fatigue level (Cornet VP, Holden R J 2018 Systematic review of smartphone-basedpassive sensing for health and wellbeing. J Biomed. Inform. 77, 120-132); the accelerometer may provide information on movement and physical activity during sleep.

1.2. ProposedService Model- PandemicPreventive Measure as a service (PPMaaS)

PPMaaS under Private Cloud offers low-cost, early detection services for COVID-19 like pandemic disease with less effort to reduce cross-infection through the Smart Screening Tool. From data collection phase to decision phase the actual physical resources invoked by the Virtual Machines in such a way that once the PPMaaS model is enrolled by a customer or cloud service provider, no possibility of direct interference between users and Cloud Service Providers (CSP), as a result to avoid being partial either or both sides.

1.3. Cloud Service Provider (CSP)

A service provider that provides some storage or software or hardware or network infrastructure to the customers through SaaS or IaaS or PaaS is called a cloud service provider which not only offers services to companies or individuals, but also ensures that it takes responsibility for anything relevant to the customers' applications.

1.4. Government Health Administration System (GHAS)

The Government can create a special unit to track and deal with the pandemic situation.

1.5. Smart ScreeningDevice (SSD)

For spot diagnosis of the disease, a small device made by sensors connected to the cloud environment can be used to sense data.

1.6. PandemicService Controller (PSC)

For maintaining a smooth contact between SSD and CSP, PSC plays a major role as it compelled to register SSD and CSP's before assigning resources to the cloud server and to avoid potential conflict.

1.7. Smart PandemicDeterminer (SPD)

After receiving the periodic signals from the CSPs, SPD will use the available cloud resources and will be able to collect data from SSD, analyze it and eventually identify the disease itself. SPD therefore maintains a mapping table of PSC to its respective CSPs', as well as a log table of SSDs' connected to it.

The increasing demand for embedded sensors coupled with pay-based infrastructures and services from Evolving Cloud Computing would make it a better technology for consistent and remote monitoring of a person's health with minimal extra costs. Consider the current scenario and based on our previous observations on various cloud management services (Bose R, Biswas H, SarddarD, Sanyal M K 2016 Cloud Billing & Verification Of Consumed Resources and Storage Spaces by a Cloud User in InternationalJournalofApplied EngineeringResearch ISSN 0973-4562 Volume 11, Number 9 pp 6568-6576) we are proposing a cloud computing-based "Pandemic Preventive Measure as a service (PPMaaS)" that accommodates people for preliminary testing as an early detection of any pandemic diseases by means of low-cost sensors with less effort to mitigate cross-infection.

Figure 2 represents our proposed "PPMaaS" service model, where PSC provides users with the services from sample selection to testing phase with proper use of cloud resources and finally submits the reports to the GHAS to take the required steps.

1.8. PPMaaS model (Working Procedure)

The proposed PPMaaS model (see figure 2.), where periodic signals, report information, and available resources sent by the CSP to SPD after effective contracts. Upon acquiring resources, SPD establishes and manages the Pandemic Cloud Service such as COVID-19 or any other, and also retains the CSPs' log table for any potential reference. At the other hand, SPD provides the Pandemic Cloud Services to the SSD via PSC after receiving requests from SSD for on-demand service, and also maintains an SSD-PSC mapping table as a guide to who and from where the service is intended. According to our suggestion, in order to disperse the infection among others, the government will take the initiation as a compulsive test for mass citizens by installing SSD into public transition areas to detect any disease outbreak stops early. After installing SSD in heavily public transfer areas (such as Railway Station, Airports, Bus Terminals, Supermarket, Shopping Mall, etc.) it must be enabled with Cloud environment through a data processing service provider (here, PSC). When a client goes to a specific place to fulfil his / her requirement, he / she must go through the screening process (by SSD) before entering the location for the preliminary ongoing pandemic detection. After receiving the client's sample data SSD sends it for analysis to the PSC. At the other hand, SPD collects data from different SSDs regularly through its FETCHING module under PSC, and TESTING module is now engaged in the identification of the appropriate tools for measuring the collected samples. DETECTION module eventually takes the decision based on the evaluated result whether the person is "Infected" or "Not Infected" and sends a message to the individual. For those infected, a message along with the current location will be sent to the nearest SHC. Ultimately, SHC periodically prepares and submits Pandemic review reports to GHAS for the required action.

Our new SCPD (Smart Cloud Pandemic Detection) algorithm handles all processes automatically and also places Virtual Machines (VMs) to the actual Physical Machines (PMs) to assign the minimum amount of resources needed. Mapping the VMs' to PMs' performed by SPD under PPMaaS. So it's really clear that no one can communicate directly with each other to access this service without either the users or the CSPs' permission from the PSC. In the case of any conflict, SPD can obtain the suspected customers' comprehensive records, or CSPs, from its own SSD-PSC as well as from the PSC-CSP Mapping log table if necessary. Thus there is no chance that Cloud users may be confused or deceived by the COSP or the CSP.

1.9. Working procedure ofSmart Pandemic Determiner (SPD) Module

Figure 3 displays the different SPD sub modules and the movement from one module to another for preliminary pandemic detection. The first module i.e., the FETCHING module collects the Client samples from the SSD via the sensors embedded in it. TESTING module is now engaged for symptom prediction following sensing of the data by the respective sensors. Within this module various algorithms are embedded to evaluate the samples to predict the symptoms. Each colored line uniquely represents the flow from sample collection to the process of symptom prediction.

Let us describe some notations used for symptom prediction by TESTING module VCB - Volumetric Changes in Blood PixelIntensity -Intensity of Pixel for each frame RGBChannel - Record video through Red-Blue-Green Channel SFFT - Simple First Fourier Transform SFFTRedChannel - PPG obtained via Red Channel compared to Blue and Green Channel using Simple Fast Fourier Transform analysis Threshold -- Difference between Maximum and Minimum Intensity

/ For Heart rate, Respiratory rateDetection Image - Captured by Camera Sensor HR- Heart Rate RR - Respiration Rate BloodVessels<- Use the Green Channel captured video P_Array - The camera sensor's Photo detector Array detects light reflection (in variations) from a Specific section of the face

//For Cough and Breathing Trouble Audio_Signal -Sound (Cough, Breathing) CBT - Cough and Breathing Trouble

/ For Cough Analysis Audio_Signal -Sound (Cough, Throat _clearing, Speech, Noise) Spectogram - Contain different audio signals of Cough, Throat Clearing, Speech and Noise PCA - Principal Component Analysis which isolate spectrogram of the cough sound from other Sound CA-Cough Analysis

//For Nasal symptoms AudioData - {Blowing the nose, sneezing, runny nose, silence, speech} ARM - Acoustic Recognition Model

CD - Classified Data SVM - Support Vector Machine Nasal- Blowing the nose, sneezing, runny nose

//For Stress level detection Conversation - Recording the conversation with Microphone S_Level - Stress Level

// ForLung Dysfunction FT - FlowTime Diagram VT - Volume _Time Diagram FV - Flow _Volume Diagram FA - Flow rate of Air FVC - Forced Vital Capacity i.e., total volume emitted during expiry FEV1 - Exhaled volume in the first second R_FVCFEV1 - Ratio of FVC and FEVI PEF - Peak Expiratory Flow i.e. maximum speed of flow velocity achieved during the test DAirflow - Degree of Airflow L_Function - Lung Dysfunction i.e., Mild, Moderate or Severe

1.10. SCPD (Smart Cloud Pandemic detection) algorithm 17. Pandemic Service Controller (PSC) offers preliminary detection of Pandemic disease using Smart Screening Device (SSD) through Smart Pandemic Determiner (SPD) in cloud environment 18. SSD may be installed in important places 19. SSD collects symptoms from the Clients 20. SSD requests to PSC for diagnosing 21. PSC checks for the SSD's authentication and Service Level Agreements (SLA) 22. If Authentic 22.1 Service accepted and acknowledgement sends to the SSD 22.2 PSC requests to the intended CSP for getting the on-demand cloud resources 22.3 CSP checks for the PSC's authentication and SLA 22.4 If bona fide 22.4.1 Service established and acceptance message sends to PSC

22.4.2 Based on the agreement CSP sends the periodic signals / resource details to Smart Pandemic Determiner (SPD) 22.4.3 SPD store the record details in a log file 22.4.4 Go to step 8 22.5 Else 22.5.1 Message - "Not acknowledged for services" back to the PSC 22.5.2 Go to step 6.2 23. Else 23.1 Message - "Not accepted for services" returns to the SSD 23.2 Go to step 4 24. SPD collects the client's symptoms from the SSD by its Fetching module 25. SPD fetches the periodic record details about the resources from its log file 26. Based on the collected symptoms SPD analysis on the data by its Testing module 26.1 If within Threshold 26.1.1 "Not Infected" -- message returned to the SSD as well as to the Client 26.1.2 Go to step 11 26.2 Else 26.2.1 SPD trace the current location of the infected person 26.2.2 "Infected" -- message returned to the SSD, to the Client, as well as to the nearest Sanatorium / Health Centre (SHC) with details about the Client 27. Check for the availability of the client 28. If client exists 28.1 Go to step 3 29. Else 13.1 Go to step 14 30. SHC sends the pandemic reports to the Government Health Management System (GHAS) 31. GHAS takes necessary actions 32. End

The end result of our approach is to create a portable testing device with an efficient interface that can be incorporated as part of the patient safety management system into medical and home equipment. Approached PPMaaS model recognizes actual physical server load next to VM user constraints addressing the mapping problem of client 's tasks into actual servers in such a way that not only reduce the number of nodes used, also the physical machines with overuse or underuse can be defined and resolved simultaneously without violating any Service Quality and Service Level Agreements. Since we consider this to be a pandemic tool that not only serves as an early detection of pandemic diseases to prevent cross-infection, also it acts as an intermediary between clients and CSPs', ensuring that nobody can communicate with each other without PSC's permission. This will eliminate the inequity between the customers 'actual usage of energy with the billing records by the suppliers', therefore avoiding any fake claims that might be brought against one another in order to get unlawful compensation.

Based on the present critical pandemic situation (COVID-19) much of the work on the COVID-19 analysis of our proposed model has been covered. The result analysis can be performed after implementation of the model. Our target is to increase the QoS with costs savings and in near future our aim is to accomplish more cloud computing capacity to battle other pandemic diseases from home as soon as possible at lower prices, less overheads.

Claims

1. An IoT-based smart pandemic detection device on a cloud-oriented architecture Pandemic Preventive Measure as a Service (PPMaaS) network for early tracking of any pandemic disease affected populations, deploying in crowd transition areas and taking precautions by contacting nearest relevant hospitals to access treatment immediately based on the condition of the patients, wherein the device based on an A enabled cloud computing based framework comprising Smart Pandemic Determiner (SPD) under Pandemic Service Controller (PSC) expertise, operates from sample collection to detection phase based on the available resources provided by the respective cloud service provide (CSP) and sends the result to the Government Health Administration System (GHAS) through nearest Sanatorium or Health Centre (SHC) for preventive measure.

2. The device as claimed in claim 1, wherein it collects data from wide range of sensors such as temperature, cameras, inertial, proximity, microphones, colour, humidity, and likes.

3. The device as claimed in claims 1 to 2, wherein the Al enabled cloud computing based framework combines the usage of the sensors with cloud environment, converted it into a smarter portable system to gain more access and analyse power for COVID-19 like pandemic detection.

4. The device as claimed in claims 1 to 3, wherein - to calculate different health issues, such as respiratory rate, heart rate, heart rate variability and health issues like skin & eye disease a advanced day-to-day sensors is used; - to track both the 'heart rate' and the 'rate of Respiration' the front and rear camera sensors of a smart phone is used and wherein 'photoplethysmogram (PPG) fingertip signal' from the top of the 'rear-facing camera' is used to achieve the heart rate; - for estimation of heart rate and heart rate variability through bare skin visual 'photoplethysmogram (PPG) signal' camera and microphone are mutually being used; - to predict fever, the sensor temperature-fingerprint is used; - to detect nasal symptoms, the application software Listen-to-Nose is used; - to test the lung function "SpiroSmart" is microphone based application software is used;

- to detect a coughing and respiratory sound microphone sensor is used; - to assess mental health and physical activity and movement, embedded sensors like accelerometer, light, GPS, microphone etc. are used.

5. The device as claimed in claims 1 to 4, wherein the Cloud Service Provider (CSP) provides storage or software or hardware or network infrastructure to the customers through SaaS or IaaS or PaaS.

6. The device as claimed in claims 1 to 5, wherein the Government can create a special unit to track and deal with the pandemic situation.

7. The device as claimed in claims 1 to 6, wherein for spot diagnosis of the disease, a small device made by sensors connected to the cloud environment is used to sense data.

8. The device as claimed in claims 1 to 7, wherein for maintaining a smooth contact between SSD and CSP, PSC plays a major role as it compelled to register SSD and CSP's before assigning resources to the cloud server and to avoid potential conflict; and wherein after receiving the periodic signals from the CSPs, SPD will use the available cloud resources and will be able to collect data from SSD, analyze it and eventually identify the disease itself.

9. The device as claimed in claims 1 to 8, where the device provides for a portable testing device with an efficient interface that can be incorporated as part of the patient safety management system into medical and home equipment

10. An IoT-based smart pandemic detection method on a cloud-oriented architecture Pandemic Preventive Measure as a Service (PPMaaS) network for early tracking of any pandemic disease affected populations, deploying in crowd transition areas and taking precautions by contacting nearest relevant hospitals to access treatment immediately based on the condition of the patients, wherein it comprises the following SCPD (Smart Cloud Pandemic detection) algorithm: 1. Pandemic Service Controller (PSC) offers preliminary detection of Pandemic disease using Smart Screening Device (SSD) through Smart Pandemic Determiner (SPD) in cloud environment 2. SSD may be installed in important places

3. SSD collects symptoms from the Clients 4. SSD requests to PSC for diagnosing 5. PSC checks for the SSD's authentication and Service Level Agreements (SLA) 6. If Authentic 6.1 Service accepted and acknowledgement sends to the SSD 6.2 PSC requests to the intended CSP for getting the on-demand cloud resources 6.3 CSP checks for the PSC's authentication and SLA 6.4 If bona fide 6.4.1 Service established and acceptance message sends to PSC 6.4.2 Based on the agreement CSP sends the periodic signals / resource details to Smart Pandemic Determiner (SPD) 6.4.3 SPD store the record details in a log file 6.4.4 Go to step 8 6.

5 Else 6.5.1 Message - "Not acknowledged for services" back to the PSC 6.5.2 Go to step 6.2 7. Else 7.1 Message - "Not accepted for services" returns to the SSD 7.2 Go to step 4 8. SPD collects the client's symptoms from the SSD by its Fetching module 9. SPD fetches the periodic record details about the resources from its log file 10. Based on the collected symptoms SPD analysis on the data by its Testing module 10.1 If within Threshold 10.1.1 "Not Infected" -- message returned to the SSD as well as to the Client 10.1.2 Go to step 11 10.2 Else 10.2.1 SPD trace the current location of the infected person 10.2.2 "Infected" -- message returned to the SSD, to the Client, as well as to the nearest Sanatorium / Health Centre (SHC) with details about the Client 11. Check for the availability of the client 12. If client exists 12.1 Go to step 3 13. Else 13.1 Go to step 14

14. SHC sends the pandemic reports to the Government Health Management System (GHAS) 15. GHAS takes necessary actions 16. End

Page 1 of 4

DRAWINGS 27 Jun 2021 2021103658

Figure 1. Proposed Service Model including basicCloud Service Models

2021103658 Page 2 of 4

Figure 2. Proposed Architecture of Pandemic Preventive Measure as a service (PPMaaS) Model

2021103658 Page 3 of 4

Figure 3. Process Flow Diagram of Smart Pandemic Determiner (SPD)

2021103658 Page 4 of 4

Figure 4. Flowchart of PPMaaS Model