CN114888870A - Insole cutting method for artificial intelligent diabetic foot ulcer identification based on thermal imaging - Google Patents
Insole cutting method for artificial intelligent diabetic foot ulcer identification based on thermal imaging Download PDFInfo
- Publication number
- CN114888870A CN114888870A CN202210217935.XA CN202210217935A CN114888870A CN 114888870 A CN114888870 A CN 114888870A CN 202210217935 A CN202210217935 A CN 202210217935A CN 114888870 A CN114888870 A CN 114888870A
- Authority
- CN
- China
- Prior art keywords
- insole
- image
- thermal imaging
- foot ulcer
- diabetic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B26—HAND CUTTING TOOLS; CUTTING; SEVERING
- B26D—CUTTING; DETAILS COMMON TO MACHINES FOR PERFORATING, PUNCHING, CUTTING-OUT, STAMPING-OUT OR SEVERING
- B26D5/00—Arrangements for operating and controlling machines or devices for cutting, cutting-out, stamping-out, punching, perforating, or severing by means other than cutting
- B26D5/007—Control means comprising cameras, vision or image processing systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0033—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0077—Devices for viewing the surface of the body, e.g. camera, magnifying lens
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
- A61B5/015—By temperature mapping of body part
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/44—Detecting, measuring or recording for evaluating the integumentary system, e.g. skin, hair or nails
- A61B5/441—Skin evaluation, e.g. for skin disorder diagnosis
- A61B5/445—Evaluating skin irritation or skin trauma, e.g. rash, eczema, wound, bed sore
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C64/00—Additive manufacturing, i.e. manufacturing of three-dimensional [3D] objects by additive deposition, additive agglomeration or additive layering, e.g. by 3D printing, stereolithography or selective laser sintering
- B29C64/30—Auxiliary operations or equipment
- B29C64/386—Data acquisition or data processing for additive manufacturing
- B29C64/393—Data acquisition or data processing for additive manufacturing for controlling or regulating additive manufacturing processes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B33—ADDITIVE MANUFACTURING TECHNOLOGY
- B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
- B33Y50/00—Data acquisition or data processing for additive manufacturing
- B33Y50/02—Data acquisition or data processing for additive manufacturing for controlling or regulating additive manufacturing processes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29L—INDEXING SCHEME ASSOCIATED WITH SUBCLASS B29C, RELATING TO PARTICULAR ARTICLES
- B29L2031/00—Other particular articles
- B29L2031/48—Wearing apparel
- B29L2031/50—Footwear, e.g. shoes or parts thereof
- B29L2031/507—Insoles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Materials Engineering (AREA)
- Chemical & Material Sciences (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Manufacturing & Machinery (AREA)
- Mechanical Engineering (AREA)
- General Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Databases & Information Systems (AREA)
- Dermatology (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Optics & Photonics (AREA)
- Forests & Forestry (AREA)
- Footwear And Its Accessory, Manufacturing Method And Apparatuses (AREA)
Abstract
The invention discloses an insole cutting method for artificial intelligence diabetic foot ulcer identification based on thermal imaging, relates to the technical field of diabetic insoles, and solves the problems that a patient at the early stage of diabetic foot ulcer cannot identify the position of the foot ulcer, the manufactured insole cannot cut the position of the ulcer, the foot skeleton of the patient is not diseased, but the stress point of the insole is possibly changed, so that the stress analysis has errors. The invention provides an insole cutting method for artificial intelligent diabetic foot ulcer identification based on thermal imaging, which is characterized in that thermal images of both feet of a diabetic are obtained by an infrared thermal imaging technology, the skin temperature condition of the foot skin caused by peripheral neurovascular lesion is displayed in a more objective thermal image form, the occurrence part and degree of the foot ulcer are further identified in an early stage, the foot lesion area in the image is intelligently identified and divided based on the result displayed by the infrared thermal imaging, a cutting template is obtained, and an individualized insole is cut, so that different requirements are met.
Description
Technical Field
The invention relates to the technical field of diabetic insoles, in particular to an insole cutting method based on thermal imaging and used for identifying artificial intelligent diabetic foot ulcer.
Background
Diabetic foot is one of the most serious complications of diabetes, and among the elderly diabetic patients, the amputated diabetic foot accounts for 50% of the non-traumatic amputated patients, wherein 85% of the amputations of diabetic foot are caused by foot ulcer and gangrene and have high infection fatality rate. Daily wear of inappropriate shoes and insoles has taken a considerable place among the many causative factors of diabetic foot ulcers. Therefore, the early identification of the diabetic foot ulcer occurrence risk area and the establishment of the personalized insole aiming at the risk area are of great significance for preventing the diabetic foot ulcer. With the increasing popularization of various diabetic foot test systems and the popularization and application of 3D printing technology in the field of medical research, individual patients are tested on feet, and personalized diabetic foot insoles are designed and customized to prevent and treat diabetic foot ulcers, so that the individual diabetic foot insoles become more and more important contents and consensus in diabetic foot prevention and treatment work. The prevention and treatment of diabetic foot ulcers by wearing personalized diabetic foot footwear is popular among patients in developed countries, but relatively few in China.
Chinese patent CN106372374B discloses a personalized design method for a diabetic foot insole, which comprises the steps of establishing a general foot skeleton model, collecting key dimension data of a patient foot, zooming the general foot skeleton model to obtain a personalized skeleton model of the patient, collecting three-dimensional data of the appearance of the outer surface of the patient foot, coupling the three-dimensional data with the personalized skeleton model, and assembling the three-dimensional data in CAD software to obtain a personalized foot three-dimensional model of the patient; then, on the basis of the three-dimensional model of the foot, according to the appearance characteristics of the sole of the foot, a diabetic foot insole basic model which is completely attached to the foot is established, the diabetic foot insole basic model and the three-dimensional model of the foot are led into finite element analysis software for assembly and grid division, a finite element model of the foot-insole is obtained, then the stress distribution result is analyzed and calculated, the modulus and the structure of each area of the diabetic foot insole basic model are modified, and the final diabetic foot insole model is formed.
Although the invention solves the problems in the background art to some extent, the following problems exist in the application: 1. for patients with early diabetic foot ulcer, the position of the foot ulcer cannot be identified, and the manufactured insole cannot cut the position of the ulcer; 2. the stress analysis is modeled according to a foot skeleton model of a patient, but for the patient with early diabetic foot ulcer, the foot skeleton is not diseased, but the stress point of the foot skeleton may change, so that the stress analysis has errors.
Disclosure of Invention
The invention aims to provide an insole cutting method for identifying artificial intelligent diabetic foot ulcers based on thermal imaging, wherein thermal images of both feet of a diabetic patient are obtained by an infrared thermal imaging technology, and compared with subjective touch, the insole cutting method can display the skin temperature condition of the foot skin caused by peripheral neurovascular lesion in a more objective thermal image form, further identify the occurrence part and degree of the foot ulcers at an early stage, intelligently identify and divide a foot lesion area in an image based on the result displayed by the infrared thermal imaging by using a computer image processing method, obtain a cutting template, cut an individualized insole and meet different requirements; the method comprises the steps of calculating upper and lower limit threshold values of pressure values in the same region by testing stress distribution conditions of double feet of a diabetic patient in multiple states, extracting corresponding lower limit threshold values of different regions, mapping the lower limit threshold values to an insole fusion image region, constructing a first layer insole model and a second layer insole model, manufacturing the first layer insole and the second layer insole by adopting a 3D printing technology, and adjusting curved surfaces of the insoles to meet requirements of different patients to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
an insole cutting method based on thermal imaging for identifying artificial intelligent diabetic foot ulcer comprises the following steps:
s1: establishing a foot ulcer risk prediction model;
s11 image acquisition phase: acquiring a large number of original diabetic foot ulcer infrared thermal imaging images, performing image enhancement processing on the acquired infrared thermal imaging images by a computer image processing method, highlighting color features of different areas in the images, deleting character annotation information in the images, only retaining original image information, performing image data set processing on all the original diabetic foot ulcer infrared thermal imaging images, diagnosing and marking the annotated acquired infrared thermal imaging images by medical personnel, sorting the foot ulcer infrared thermal imaging images of all patients, confirming image quality by the medical personnel, and determining ulcer areas in the images;
s12 "gold standard" image confirmation stage: finding the most clear and obvious representative image from all infrared thermal imaging images, manually deleting the ulcer area in the image by medical staff, and making a 'gold standard' image as an evaluation standard for constructing a prediction model by a computer;
s13 image dataset preparation phase: normalizing all the infrared thermal imaging images, and marking normal images and ulcer images by medical staff;
s14 image risk prediction model construction stage: after the diabetic foot ulcer infrared thermal imaging image data set is obtained, different 'deep learning' methods are selected, and a diabetic foot ulcer infrared thermal imaging image risk prediction model based on a deep convolutional neural network is constructed;
s2: preliminarily generating an insole template;
s21 preliminary forming stage: the foot ulcer risk prediction model identifies a plantar ulcer area and a plantar normal skin area in the diabetic foot ulcer infrared thermal imaging image, realizes semantic segmentation of the two areas, filters the plantar ulcer area and leaves an infrared thermal imaging image of the plantar normal skin of a patient;
s22 insole matching stage: confirming standard size insoles of both feet of a patient, and shooting insole images by using a normal camera;
s23 insole synthesis stage: and selecting a proper computer image fusion algorithm, and carrying out an image fusion experiment on the two images to finally obtain a template image cut by the insole.
Preferably, the specific method for constructing the image risk prediction model stage comprises the following steps:
s141: constructing an image classification and identification model of a deep learning method, selecting a deep convolution neural network structure to train a diabetic foot ulcer infrared thermal imaging image data set, completing image classification, and finally obtaining a risk prediction model capable of correctly distinguishing normal plantar infrared thermal imaging images and ulcer infrared thermal imaging images;
s142: an image semantic segmentation model of a deep learning method is constructed, an image data set is loaded into the model to be trained, semantic segmentation of the infrared thermal imaging image of the sole is realized, a normal sole region and a plantar ulcer region in the infrared thermal imaging image are identified, and semantic segmentation of the two regions is realized.
Preferably, the insole synthesis stage specific method comprises the following steps:
s231: firstly, obtaining the whole sole area of an infrared thermal imaging image A by using an area growing method, and projecting the area onto a visible light insole image B;
s232: respectively carrying out combined complex shear wave transformation decomposition on the infrared thermal imaging image A and the visible light insole image B to obtain multi-scale decomposition coefficients of the infrared thermal imaging image A and the visible light insole image B, wherein the multi-scale decomposition coefficients comprise a high-frequency sub-band coefficient and a low-frequency sub-band coefficient;
s233: selecting different fusion strategies for the regions with different decomposition coefficients to obtain high-frequency and low-frequency fusion coefficients;
s234: and performing combined complex shear wave inverse transformation on the fused coefficients to reconstruct a fused image.
Preferably, the insole cutting method based on thermal imaging for identifying artificial intelligent diabetic foot ulcers further comprises a step S3 of adjusting an insole template, wherein the step of adjusting the insole template specifically comprises the following steps:
s31 pressure value acquisition stage: projecting the fused image onto a pressure sensing flat plate, stepping a foot of a diabetic on the pressure sensing flat plate to correspond to the insole fused image, and acquiring a pressure value of the insole fused image area by the pressure sensing flat plate;
s32 pressure value optimizing stage: determining different stress thicknesses of the insole of the diabetic patient according to the stress positions of the sole of the patient, and making a first layer insole model and a second layer insole model;
s33 stage of synthesizing insole template: and the first layer insole model and the second layer insole model are overlapped to obtain a final insole template.
Preferably, the specific method for optimizing the pressure value stage comprises the following steps:
s321: collecting pressure values of the diabetic patients for multiple times, converging the pressure values into an exclusive data set of the diabetic patients, carrying out regional division on stress positions on the fused image, and summarizing the pressure values in the same region;
s322: calculating upper and lower limit threshold values of a pressure value in the same region, extracting corresponding lower limit threshold values of different regions, mapping the lower limit threshold values to an insole fusion image region, reducing the thickness of a diabetic foot ulcer insole template in a high pressure region and increasing the thickness of the diabetic foot ulcer insole template in a low pressure region according to the extracted pressure data to optimize the contact area of the two, and generating a first layer insole model;
s323: and calculating the pressed interval values corresponding to different areas, mapping the pressed interval values to the insole fusion image area, reducing the thickness of the diabetic foot ulcer insole template in the bottom pressed interval value area according to the extracted pressed interval values, increasing the thickness of the diabetic foot ulcer insole template in the high-bottom pressed interval value area, and generating a second-layer insole model.
Preferably, the exclusive data set comprises stress distribution of both feet of the diabetic patient when standing, stress distribution of both feet of the diabetic patient when sitting or standing, and stress distribution of both feet of the diabetic patient when walking.
Preferably, the adjusting insole template further comprises an S34 insole making stage:
s341: printing a first layer of insole and a second layer of insole according to the first layer of insole model and the second layer of insole model by a 3D printing technology;
s342: the first layer insole and the second layer insole are connected by gluing or sewing.
Preferably, the second layer of insoles are made of sponge materials, and the hardness of the first layer of insoles is greater than that of the second layer of insoles.
Preferably, the pressure sensing panel comprises a device body, a display screen, a piezoelectric film sensor, a data transmission module and a model generation module, wherein the display screen is arranged on the surface of the device body, the piezoelectric film sensor is arranged on the lower surface of the display screen, the piezoelectric film sensor is electrically connected with the data transmission module, and the data transmission module is electrically connected with the model generation module.
Preferably, the piezoelectric film sensors are uniformly distributed in a block shape.
Compared with the prior art, the invention has the beneficial effects that: the invention provides an insole cutting method for identifying artificial intelligent diabetic foot ulcers based on thermal imaging, which is characterized in that thermal images of both feet of a diabetic patient are obtained by an infrared thermal imaging technology, compared with subjective touch, the skin temperature condition of the skin of the foot caused by peripheral neurovascular lesion can be displayed in a more objective thermal image form, the occurrence part and degree of the foot ulcers can be further identified at an early stage, the foot lesion area in the image is intelligently identified and divided by a computer image processing method based on the result displayed by the infrared thermal imaging, a cutting template is obtained, and an individualized insole is cut to meet different requirements; the method comprises the steps of calculating upper and lower limit threshold values of pressure values in the same region by testing the stress distribution conditions of the two feet of a diabetic patient in multiple states, extracting the lower limit threshold values corresponding to different regions, mapping the lower limit threshold values to an insole fusion image region, constructing a first layer insole model and a second layer insole model, manufacturing the first layer insole and the second layer insole by adopting a 3D printing technology, and adjusting the curved surfaces of the insoles to adapt to the requirements of different patients.
Drawings
FIG. 1 is a flow chart of the present invention for establishing a model for predicting risk of foot ulcer;
FIG. 2 is a flow diagram of a preliminary generation insole template of the present invention;
FIG. 3 is a flowchart of the stages of constructing an image risk prediction model according to the present invention;
FIG. 4 is an algorithm diagram of the insole synthesis phase of the present invention;
FIG. 5 is a flow chart of an embodiment of the present invention;
FIG. 6 is a flow chart of the present invention for adjusting the insole template;
FIG. 7 is a flow chart of the pressure value optimization stage of the present invention;
FIG. 8 is a diagram of a pressure sensing plate according to the present invention;
fig. 9 is a partial view of the pressure sensing plate of the present invention.
In the figure: 1. a device body; 2. a display screen; 3. a piezoelectric film sensor; 4. a data transmission module; 5. and a model generation module.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The first embodiment is as follows:
referring to fig. 1-4, the method for cutting out an insole based on thermal imaging artificial intelligence diabetic foot ulcer identification comprises the following steps:
s1: establishing a foot ulcer risk prediction model;
s11 image acquisition phase: acquiring a large number of original diabetic foot ulcer infrared thermal imaging images, performing image enhancement processing on the acquired infrared thermal imaging images by a computer image processing method, highlighting color features of different areas in the images, deleting character annotation information in the images, only retaining original image information, performing image data set processing on all the original diabetic foot ulcer infrared thermal imaging images, diagnosing and marking the annotated acquired infrared thermal imaging images by medical personnel, sorting the foot ulcer infrared thermal imaging images of all patients, confirming image quality by the medical personnel, and determining ulcer areas in the images;
s12 "gold standard" image confirmation stage: finding the most clear and obvious representative image from all infrared thermal imaging images, manually deleting the ulcer area in the image by medical staff, and making a 'gold standard' image as an evaluation standard for constructing a prediction model by a computer;
s13 image dataset preparation phase: normalizing all the infrared thermal imaging images, and marking normal images and ulcer images by medical staff;
s14 image risk prediction model construction stage: after the infrared thermal imaging image data set of the diabetic foot ulcer is obtained, different 'deep learning' methods are selected, a diabetic foot ulcer infrared thermal imaging image risk prediction model based on a deep convolutional neural network is constructed, and the specific method for constructing the image risk prediction model stage comprises the following steps:
s141: constructing an image classification and identification model of a deep learning method, selecting a deep convolution neural network structure to train a diabetic foot ulcer infrared thermal imaging image data set, completing image classification, and finally obtaining a risk prediction model capable of correctly distinguishing normal plantar infrared thermal imaging images and ulcer infrared thermal imaging images;
s142: constructing a 'deep learning method' image semantic segmentation model, loading an image data set into the model for training, realizing semantic segmentation of the infrared thermal imaging image of the sole, identifying a normal sole region and a plantar ulcer region in the infrared thermal imaging image, and realizing semantic segmentation of the two regions
S2: preliminarily generating an insole template;
s21 preliminary forming stage: the foot ulcer risk prediction model identifies a plantar ulcer area and a plantar normal skin area in the diabetic foot ulcer infrared thermal imaging image, realizes semantic segmentation of the two areas, filters the plantar ulcer area and leaves an infrared thermal imaging image of the plantar normal skin of a patient;
s22 insole matching stage: confirming standard-sized insoles of both feet of a patient, and shooting insole images by using a normal camera to ensure that the shot insole images are the same as the infrared thermal imaging images of the diabetic foot ulcers;
s23 insole synthesis stage: selecting a proper computer image fusion algorithm, carrying out an image fusion experiment on the two images to finally obtain a template image clipped by the insole, obtaining the template image clipped by the insole by fusing the images, and determining a proper image fusion method through continuous preferential experiments when the experiment result is not ideal enough to further obtain a more practical clipping template, wherein the specific method in the insole synthesis stage comprises the following steps:
s231: firstly, obtaining the whole sole area of an infrared thermal imaging image A by using an area growing method, and projecting the area onto a visible light insole image B;
s232: respectively carrying out combined complex shear wave transformation decomposition on the infrared thermal imaging image A and the visible light insole image B to obtain multi-scale decomposition coefficients of the infrared thermal imaging image A and the visible light insole image B, wherein the multi-scale decomposition coefficients comprise a high-frequency sub-band coefficient and a low-frequency sub-band coefficient;
s233: selecting different fusion strategies for the regions with different decomposition coefficients to obtain high-frequency and low-frequency fusion coefficients;
s234: and performing combined complex shear wave inverse transformation on the fused coefficients to reconstruct a fused image.
In this example, the specific method is as follows, for each patient: shooting a plantar infrared thermal imaging image of a patient to be detected by using an infrared camera; preprocessing the obtained image by a computer, wherein the image processing comprises image enhancement, irrelevant information filtering and the like; loading the processed sole infrared thermal imaging image into a diabetic foot ulcer infrared thermal imaging risk prediction model; obtaining a lesion position of the plantar infrared thermal imaging image and a plantar infrared thermal imaging image of a lesion excision position; selecting a size template image of the feet of the patient according to the standard shoe size; carrying out computer image fusion processing on the sole infrared thermal imaging image after the disease displacement is segmented and the template to obtain a template image cut by the insole; printing the fused image, and cutting the insole according to the printed image.
The construction of a diabetic foot ulcer risk prediction model is completed by analyzing the image and the temperature index by using a computer deep learning method, the identification of the infrared thermal imaging image of the diabetic foot ulcer is realized, and then a personalized diabetic foot insole is designed and customized to prevent and treat the diabetic foot ulcer.
Example two:
referring to fig. 5-9, on the basis of the completion of the insole synthesis stage of the embodiment, the present embodiment further adds a process of adjusting insole templates, wherein the process of adjusting insole templates specifically includes the following steps:
s31 pressure value acquisition stage: projecting the fused image onto a pressure sensing flat plate, enabling a diabetic patient to step on the pressure sensing flat plate to correspond to the insole fused image, enabling the pressure sensing flat plate to obtain a pressure value of the insole fused image area, enabling the pressure sensing flat plate to comprise a device body 1, a display screen 2, a piezoelectric film sensor 3, a data transmission module 4 and a model generation module 5, enabling the display screen 2 to be arranged on the surface of the device body 1, enabling the piezoelectric film sensor 3 to be arranged on the lower surface of the display screen 2, enabling the piezoelectric film sensor 3 to be electrically connected with the data transmission module 4, enabling the data transmission module 4 to be electrically connected with the model generation module 5, enabling the display screen 2 to be used for projecting the insole fused image, enabling the piezoelectric film sensor 3 to collect foot stress distribution of the patient, enabling the data transmission module 4 to convert and transmit data collected by the piezoelectric film sensor 3, and enabling the model generation module 5 to perform model generation according to the pressure value distribution, the piezoelectric film sensors 3 are uniformly distributed in a blocky manner, so that the stress distribution of the feet of the patient can be comprehensively collected;
s32 pressure value optimizing stage: according to the stress position of the sole of a patient, different stress thicknesses of the insole of the diabetic patient are determined, a first layer insole model and a second layer insole model are formulated, and the specific pressure value optimizing stage method comprises the following steps:
s321: collecting pressure values of the diabetic patients for multiple times, converging the pressure values into an exclusive data set of the diabetic patients, carrying out regional division on stress positions on the fused image, summarizing the pressure values in the same region, wherein the exclusive data set comprises stress distribution of both feet of the diabetic patients when the diabetic patients stand, stress distribution of both feet of the diabetic patients when the diabetic patients sit and stand and stress distribution of both feet of the diabetic patients when the diabetic patients walk;
s322: calculating upper and lower limit thresholds of pressure values in the same region, extracting corresponding lower limit thresholds of the pressure values in different regions, mapping the lower limit thresholds to the insole fusion image region, reducing the thickness of the diabetic foot ulcer insole template in the high pressure region and increasing the thickness of the diabetic foot ulcer insole template in the low pressure region according to the extracted pressure data to optimize the contact area of the two, and generating a first layer insole model;
s323: calculating the pressed interval values corresponding to different areas, mapping the pressed interval values to the insole fusion image area, reducing the thickness of the diabetic foot ulcer insole template in the bottom pressed interval value area according to the extracted pressed interval values, increasing the thickness of the diabetic foot ulcer insole template in the high-bottom pressed interval value area, and generating a second-layer insole model;
s33 stage of synthesizing insole template: the first layer insole model and the second layer insole model are overlapped to obtain a final insole template;
s34 insole making stage:
s341: printing a first layer of insole and a second layer of insole according to the first layer of insole model and the second layer of insole model by a 3D printing technology;
s342: first layer shoe-pad and second layer shoe-pad pass through the glue bonding or the mode of sewing up and connect, the second layer shoe-pad is made for sponge material, and the hardness of first layer shoe-pad is greater than the hardness of second layer shoe-pad, and when the shoe-pad was used to the diabetes mellitus patient, the second layer shoe-pad can take place deformation according to the change of patient's foot atress for the whole and patient's of shoe-pad foot laminating is inseparable.
In the embodiment, the step of adjusting the insole template is added, the distribution of the stress on the sole of the patient is increased on the original insole fusion image, the diabetic foot ulcer insole which is more tightly fitted is customized for the patient, and the overall comfort is better than that of the insole in the first embodiment.
In summary, the following steps: according to the method for cutting the insole based on the artificial intelligent diabetic foot ulcer identification based on the thermal imaging, thermal images of both feet of a diabetic patient are obtained by an infrared thermal imaging technology, compared with subjective touch, the skin temperature condition of the foot skin caused by peripheral neurovascular lesion can be displayed in a more objective thermal image form, the occurrence part and degree of the foot ulcer can be further identified in an early stage, the foot lesion area in the image can be intelligently identified and divided by a computer image processing method based on the result displayed by the infrared thermal imaging, a cutting template can be obtained, and the personalized insole can be cut to meet different requirements; the method comprises the steps of calculating upper and lower limit threshold values of pressure values in the same region by testing the stress distribution conditions of the two feet of a diabetic patient in multiple states, extracting the lower limit threshold values corresponding to different regions, mapping the lower limit threshold values to an insole fusion image region, constructing a first layer insole model and a second layer insole model, manufacturing the first layer insole and the second layer insole by adopting a 3D printing technology, and adjusting the curved surfaces of the insoles to adapt to the requirements of different patients.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be able to cover the technical solutions and the inventive concepts of the present invention within the technical scope of the present invention.
Claims (10)
1. An insole cutting method based on thermal imaging for identifying artificial intelligent diabetic foot ulcer is characterized by comprising the following steps:
s1: establishing a foot ulcer risk prediction model;
s11 image acquisition phase: acquiring a large number of original diabetic foot ulcer infrared thermal imaging images, performing image enhancement processing on the acquired infrared thermal imaging images by a computer image processing method, highlighting color features of different areas in the images, deleting character annotation information in the images, only retaining original image information, performing image data set processing on all the original diabetic foot ulcer infrared thermal imaging images, diagnosing and marking the annotated acquired infrared thermal imaging images by medical personnel, sorting the foot ulcer infrared thermal imaging images of all patients, confirming image quality by the medical personnel, and determining ulcer areas in the images;
s12 "gold standard" image confirmation stage: finding the most clear and obvious representative image from all infrared thermal imaging images, manually deleting the ulcer area in the image by medical staff, and making a 'gold standard' image as an evaluation standard for constructing a prediction model by a computer;
s13 image dataset preparation phase: normalizing all the infrared thermal imaging images, and marking normal images and ulcer images by medical staff;
s14 image risk prediction model construction stage: after the diabetic foot ulcer infrared thermal imaging image data set is obtained, different 'deep learning' methods are selected, and a diabetic foot ulcer infrared thermal imaging image risk prediction model based on a deep convolutional neural network is constructed;
s2: preliminarily generating an insole template;
s21 preliminary forming stage: the foot ulcer risk prediction model identifies a plantar ulcer area and a plantar normal skin area in the diabetic foot ulcer infrared thermal imaging image, realizes semantic segmentation of the two areas, filters the plantar ulcer area and leaves an infrared thermal imaging image of the plantar normal skin of a patient;
s22 insole matching stage: confirming standard size insoles of both feet of a patient, and shooting an insole image by using a normal camera;
s23 insole synthesis stage: and selecting a proper computer image fusion algorithm, and carrying out an image fusion experiment on the two images to finally obtain a template image cut by the insole.
2. The method of claim 1, wherein the insole is cut by artificial intelligence based on thermal imaging for diabetic foot ulcer identification, and the method comprises: the specific method for constructing the image risk prediction model stage comprises the following steps:
s141: constructing an image classification and identification model of a deep learning method, selecting a deep convolution neural network structure to train a diabetic foot ulcer infrared thermal imaging image data set, completing image classification, and finally obtaining a risk prediction model capable of correctly distinguishing normal plantar infrared thermal imaging images and ulcer infrared thermal imaging images;
s142: an image semantic segmentation model of a deep learning method is constructed, an image data set is loaded into the model to be trained, semantic segmentation of the infrared thermal imaging image of the sole is realized, a normal sole region and a plantar ulcer region in the infrared thermal imaging image are identified, and semantic segmentation of the two regions is realized.
3. The method of claim 1, wherein the insole is cut by artificial intelligence based on thermal imaging for diabetic foot ulcer identification, and the method comprises: the specific method of the insole synthesis stage comprises the following steps:
s231: firstly, obtaining the whole sole area of an infrared thermal imaging image A by using an area growing method, and projecting the area onto a visible light insole image B;
s232: respectively carrying out combined complex shear wave transformation decomposition on the infrared thermal imaging image A and the visible light insole image B to obtain multi-scale decomposition coefficients of the infrared thermal imaging image A and the visible light insole image B, wherein the multi-scale decomposition coefficients comprise a high-frequency sub-band coefficient and a low-frequency sub-band coefficient;
s233: selecting different fusion strategies for the regions with different decomposition coefficients to obtain high-frequency and low-frequency fusion coefficients;
s234: and performing combined complex shear wave inverse transformation on the fused coefficients to reconstruct a fused image.
4. The method of claim 1, further comprising the step of S3 adjusting insole template, wherein the method comprises: the insole template adjustment method specifically comprises the following steps:
s31 pressure value acquisition stage: projecting the fused image onto a pressure sensing flat plate, stepping a foot of a diabetic on the pressure sensing flat plate to correspond to the insole fused image, and acquiring a pressure value of the insole fused image area by the pressure sensing flat plate;
s32 pressure value optimizing stage: determining different stress thicknesses of the insole of the diabetic patient according to the stress positions of the sole of the patient, and making a first layer insole model and a second layer insole model;
s33 stage of synthesizing insole template: the first layer insole model and the second layer insole model are overlapped to obtain a final insole template.
5. The method of claim 4 for insole tailoring for thermal imaging based artificial intelligence diabetic foot ulcer identification, wherein: the specific pressure value optimizing stage method comprises the following steps:
s321: collecting pressure values of the diabetic patients for multiple times, converging the pressure values into an exclusive data set of the diabetic patients, carrying out regional division on stress positions on the fused image, and summarizing the pressure values in the same region;
s322: calculating upper and lower limit thresholds of pressure values in the same region, extracting corresponding lower limit thresholds of the pressure values in different regions, mapping the lower limit thresholds to the insole fusion image region, reducing the thickness of the diabetic foot ulcer insole template in the high pressure region and increasing the thickness of the diabetic foot ulcer insole template in the low pressure region according to the extracted pressure data to optimize the contact area of the two, and generating a first layer insole model;
s323: and calculating the pressed interval values corresponding to different areas, mapping the pressed interval values to the insole fusion image area, reducing the thickness of the diabetic foot ulcer insole template in the bottom pressed interval value area according to the extracted pressed interval values, increasing the thickness of the diabetic foot ulcer insole template in the high-bottom pressed interval value area, and generating a second-layer insole model.
6. The method of claim 5, wherein the insole is cut by artificial intelligence based on thermal imaging for diabetic foot ulcer identification, and the method comprises: the exclusive data set comprises stress distribution of both feet of the diabetic patient when standing, stress distribution of both feet of the diabetic patient when sitting and standing, and stress distribution of both feet of the diabetic patient when walking.
7. The method of claim 6, wherein the insole is cut by artificial intelligence based on thermal imaging for diabetic foot ulcer identification, and the method comprises: the adjusting insole template further comprises an S34 insole making stage:
s341: printing a first layer of insole and a second layer of insole according to the first layer of insole model and the second layer of insole model by a 3D printing technology;
s342: the first layer insole and the second layer insole are connected by gluing or sewing.
8. The method of claim 7, wherein the insole is cut by artificial intelligence based on thermal imaging for diabetic foot ulcer identification, and the method comprises: the second layer of insoles are made of sponge materials, and the hardness of the first layer of insoles is greater than that of the second layer of insoles.
9. The method of claim 5 for insole tailoring based on thermal imaging for artificial intelligence diabetic foot ulcer identification, wherein: the pressure sensing flat plate comprises a device body (1), a display screen (2), a piezoelectric film sensor (3), a data transmission module (4) and a model generation module (5), wherein the display screen (2) is arranged on the surface of the device body (1), the piezoelectric film sensor (3) is arranged on the lower surface of the display screen (2), the piezoelectric film sensor (3) is electrically connected with the data transmission module (4), and the data transmission module (4) is electrically connected with the model generation module (5).
10. The method of claim 9 for insole tailoring based on thermal imaging for artificial intelligence diabetic foot ulcer identification, wherein: the piezoelectric film sensors (3) are uniformly distributed in a block shape.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210217935.XA CN114888870B (en) | 2022-03-08 | 2022-03-08 | Insole cutting method for identifying artificial intelligent diabetic foot ulcers based on thermal imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210217935.XA CN114888870B (en) | 2022-03-08 | 2022-03-08 | Insole cutting method for identifying artificial intelligent diabetic foot ulcers based on thermal imaging |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114888870A true CN114888870A (en) | 2022-08-12 |
| CN114888870B CN114888870B (en) | 2023-08-29 |
Family
ID=82714766
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210217935.XA Active CN114888870B (en) | 2022-03-08 | 2022-03-08 | Insole cutting method for identifying artificial intelligent diabetic foot ulcers based on thermal imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114888870B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116548926A (en) * | 2023-06-26 | 2023-08-08 | 中科(安徽)G60智慧健康创新研究院 | Cold stimulation-based diabetic foot screening system and method |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR200281342Y1 (en) * | 2002-04-03 | 2002-07-13 | 윤효진 | Functional Shoes Insoles |
| US20080109183A1 (en) * | 2006-11-06 | 2008-05-08 | Colorado Seminary, Which Owns And Operates The University Of Denver | Smart Insole for Diabetic Patients |
| TW200919210A (en) * | 2007-07-18 | 2009-05-01 | Steven Kays | Adaptive electronic design |
| US20120245403A1 (en) * | 2010-04-20 | 2012-09-27 | Bioelectronics Corp. | Insole Electromagnetic Therapy Device |
| CN104212174A (en) * | 2013-06-03 | 2014-12-17 | 上海申精化学科技有限公司 | Foaming material for preparing functional insole and functional insole |
| US20170127999A1 (en) * | 2015-11-06 | 2017-05-11 | Podimetrics, Inc. | Footwear System for Ulcer or Pre-Ulcer Detection |
| US20170185283A1 (en) * | 2015-12-28 | 2017-06-29 | Dexcom, Inc. | Wearable apparatus for continuous blood glucose monitoring |
| CN107019289A (en) * | 2016-08-15 | 2017-08-08 | 深圳市金火鸟全息能量科技发展连锁有限公司 | Shoe-pad and its manufacture craft |
| CN209391175U (en) * | 2018-10-12 | 2019-09-17 | 深圳市千帆电子有限公司 | A kind of insole preventing diabetes |
| CN209673265U (en) * | 2019-01-30 | 2019-11-22 | 苏州大学 | A kind of wireless pressure sensor |
| CN112100853A (en) * | 2020-09-16 | 2020-12-18 | 河海大学常州校区 | Insole personalized design method for supporting and pressure reducing functions |
| US20210046356A1 (en) * | 2015-06-23 | 2021-02-18 | Ipcomm Llc | Method and Apparatus for Early Detection of Diabetic Foot Disorders by Analyzing Foot Temperature and Vertical and Shear Forces on Feet |
| CN112989669A (en) * | 2021-04-01 | 2021-06-18 | 南通大学 | Personalized design method for diabetic foot pad |
-
2022
- 2022-03-08 CN CN202210217935.XA patent/CN114888870B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR200281342Y1 (en) * | 2002-04-03 | 2002-07-13 | 윤효진 | Functional Shoes Insoles |
| US20080109183A1 (en) * | 2006-11-06 | 2008-05-08 | Colorado Seminary, Which Owns And Operates The University Of Denver | Smart Insole for Diabetic Patients |
| TW200919210A (en) * | 2007-07-18 | 2009-05-01 | Steven Kays | Adaptive electronic design |
| US20120245403A1 (en) * | 2010-04-20 | 2012-09-27 | Bioelectronics Corp. | Insole Electromagnetic Therapy Device |
| CN104212174A (en) * | 2013-06-03 | 2014-12-17 | 上海申精化学科技有限公司 | Foaming material for preparing functional insole and functional insole |
| US20210046356A1 (en) * | 2015-06-23 | 2021-02-18 | Ipcomm Llc | Method and Apparatus for Early Detection of Diabetic Foot Disorders by Analyzing Foot Temperature and Vertical and Shear Forces on Feet |
| US20170127999A1 (en) * | 2015-11-06 | 2017-05-11 | Podimetrics, Inc. | Footwear System for Ulcer or Pre-Ulcer Detection |
| US20170185283A1 (en) * | 2015-12-28 | 2017-06-29 | Dexcom, Inc. | Wearable apparatus for continuous blood glucose monitoring |
| US20170185284A1 (en) * | 2015-12-28 | 2017-06-29 | Dexcom, Inc. | Wearable apparatus for continuous blood glucose monitoring |
| CN107019289A (en) * | 2016-08-15 | 2017-08-08 | 深圳市金火鸟全息能量科技发展连锁有限公司 | Shoe-pad and its manufacture craft |
| CN209391175U (en) * | 2018-10-12 | 2019-09-17 | 深圳市千帆电子有限公司 | A kind of insole preventing diabetes |
| CN209673265U (en) * | 2019-01-30 | 2019-11-22 | 苏州大学 | A kind of wireless pressure sensor |
| CN112100853A (en) * | 2020-09-16 | 2020-12-18 | 河海大学常州校区 | Insole personalized design method for supporting and pressure reducing functions |
| CN112989669A (en) * | 2021-04-01 | 2021-06-18 | 南通大学 | Personalized design method for diabetic foot pad |
Non-Patent Citations (3)
| Title |
|---|
| 李可心;杜玉茗;段文燕;翟英慧;肖燕;封岩;: "我国糖尿病足分级诊疗模式探讨", 临床荟萃, no. 02 * |
| 罗颖琪;李炳辉;许樟荣;SICCO A.BUS;LARRY A.LAVERY;MATILDE MONTEIRO-SOARES;ANNE RASMUSSEN;ANITA RASPOVIC;ISABEL C.N.SACCO;JAAP J.VAN NETTEN;: "国际糖尿病足工作组：糖尿病足溃疡预防指南――《国际糖尿病足工作组：糖尿病足防治国际指南（2019）》的一部分", 感染、炎症、修复, no. 03 * |
| 许娇;苏继亮;胡婷婷;洪伊荣;何朝珠;: "糖尿病足溃疡复发因素及预防策略的研究进展", 护理学杂志, no. 11 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116548926A (en) * | 2023-06-26 | 2023-08-08 | 中科(安徽)G60智慧健康创新研究院 | Cold stimulation-based diabetic foot screening system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114888870B (en) | 2023-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Adegun et al. | Deep learning-based system for automatic melanoma detection | |
| CN106372374B (en) | A kind of personalized designs method for diabetes insole | |
| CN106618481A (en) | Traditional Chinese medicine intelligent diagnosis expert system | |
| CN110875092B (en) | Health big data service method and system based on remote fundus screening | |
| CN107527326A (en) | A kind of wheat scab diagnostic method based on high light spectrum image-forming | |
| CN115131642B (en) | Multi-modal medical data fusion system based on multi-view subspace clustering | |
| CN205215184U (en) | Traditional chinese medical science intelligent diagnosis expert system | |
| CN107862249A (en) | A kind of bifurcated palm grain identification method and device | |
| CN111544256A (en) | Brain-controlled intelligent full limb rehabilitation method based on graph convolution and transfer learning | |
| CN114888870B (en) | Insole cutting method for identifying artificial intelligent diabetic foot ulcers based on thermal imaging | |
| CN114565613B (en) | Pancreas postoperative diabetes prediction system based on there is study of supervision degree of depth subspace | |
| CN106778047A (en) | A kind of traditional Chinese medical science facial diagnosis integrated system based on various dimensions medical image | |
| CN115089139A (en) | Personalized physiological parameter measuring method combining biological characteristic identification | |
| CN114550299A (en) | System and method for evaluating daily life activity ability of old people based on video | |
| CN111329467A (en) | Heart disease auxiliary detection method based on artificial intelligence | |
| CN113768471A (en) | Parkinson disease auxiliary diagnosis system based on gait analysis | |
| CN116664966B (en) | Infrared image processing system | |
| CN105708441A (en) | Wearable finger stall and electrocardiogram and pulse condition acquisition terminal | |
| CN109711306B (en) | Method and equipment for obtaining facial features based on deep convolutional neural network | |
| CN116363145A (en) | Fundus OCT image retina layering method based on double edge representation | |
| CN116543455A (en) | Method, equipment and medium for establishing parkinsonism gait damage assessment model and using same | |
| CN115409812A (en) | CT image automatic classification method based on fusion time attention mechanism | |
| Jayachandran et al. | Retinal vessels segmentation of colour fundus images using two stages cascades convolutional neural networks | |
| CN114983341A (en) | Multi-modal feature fusion based multi-classification prediction system for Alzheimer's disease | |
| Mu et al. | Automated nonlinear feature generation and classification of foot pressure lesions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |