CN113326745A - Application system for judging and identifying stoma situation through image identification technology - Google Patents
Application system for judging and identifying stoma situation through image identification technology Download PDFInfo
- Publication number
- CN113326745A CN113326745A CN202110524013.9A CN202110524013A CN113326745A CN 113326745 A CN113326745 A CN 113326745A CN 202110524013 A CN202110524013 A CN 202110524013A CN 113326745 A CN113326745 A CN 113326745A
- Authority
- CN
- China
- Prior art keywords
- module
- stoma
- image
- data
- patient
- 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.)
- Pending
Links
- 238000005516 engineering process Methods 0.000 title claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 41
- 230000006870 function Effects 0.000 claims abstract description 22
- 230000000474 nursing effect Effects 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000011282 treatment Methods 0.000 claims abstract description 12
- 238000012549 training Methods 0.000 claims description 39
- 238000004891 communication Methods 0.000 claims description 30
- 238000004140 cleaning Methods 0.000 claims description 22
- 238000004458 analytical method Methods 0.000 claims description 19
- 238000001914 filtration Methods 0.000 claims description 19
- 238000007726 management method Methods 0.000 claims description 19
- 238000012216 screening Methods 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 17
- 230000006698 induction Effects 0.000 claims description 14
- 238000013500 data storage Methods 0.000 claims description 10
- 238000013135 deep learning Methods 0.000 claims description 10
- 238000013441 quality evaluation Methods 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 8
- 239000003086 colorant Substances 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000001939 inductive effect Effects 0.000 claims description 6
- 208000024891 symptom Diseases 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000009430 construction management Methods 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 3
- 238000001303 quality assessment method Methods 0.000 claims description 3
- 230000006866 deterioration Effects 0.000 abstract description 3
- 230000008569 process Effects 0.000 abstract description 3
- 238000004590 computer program Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 230000006872 improvement Effects 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 201000004624 Dermatitis Diseases 0.000 description 2
- 208000032843 Hemorrhage Diseases 0.000 description 2
- 206010019909 Hernia Diseases 0.000 description 2
- 208000031481 Pathologic Constriction Diseases 0.000 description 2
- 208000012287 Prolapse Diseases 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 208000035475 disorder Diseases 0.000 description 2
- 230000000968 intestinal effect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000010008 shearing Methods 0.000 description 2
- 230000036262 stenosis Effects 0.000 description 2
- 208000037804 stenosis Diseases 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000011269 treatment regimen Methods 0.000 description 2
- 206010012434 Dermatitis allergic Diseases 0.000 description 1
- 206010016936 Folliculitis Diseases 0.000 description 1
- 206010017533 Fungal infection Diseases 0.000 description 1
- 208000031888 Mycoses Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 210000001367 artery Anatomy 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 201000008937 atopic dermatitis Diseases 0.000 description 1
- 208000010668 atopic eczema Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000740 bleeding effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 210000002249 digestive system Anatomy 0.000 description 1
- 208000010643 digestive system disease Diseases 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 230000002550 fecal effect Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 208000001875 irritant dermatitis Diseases 0.000 description 1
- 230000000302 ischemic effect Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 210000004877 mucosa Anatomy 0.000 description 1
- 230000017074 necrotic cell death Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 206010040882 skin lesion Diseases 0.000 description 1
- 231100000444 skin lesion Toxicity 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
- 208000014001 urinary system disease Diseases 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Abstract
The invention relates to the technical field of image recognition, in particular to an application system for judging and recognizing a stoma condition through an image recognition technology. The system comprises a system base unit, an identification management unit and a function application unit; the system basic unit is used for providing various equipment devices, technologies, applications and the like supporting system operation; the identification management unit is used for acquiring image information of the stoma and carrying out a series of processing operations on the image to obtain an identification result; the function application unit is used for adding various extended services on the basis of image recognition to complete the functions of the system. The invention can process and identify the image information of the stoma condition, evaluate the stoma condition in time, and provide nursing guidance suggestions and treatment schemes, and can provide a channel for patients to see a doctor on the internet, so that the patients can judge the stoma condition at home, thereby facilitating better nursing the stoma, reducing the probability of deterioration of the stoma, avoiding mistreatment opportunity, and reducing the economic burden caused by frequent doctor-seeing of the patients.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to an application system for judging and recognizing a stoma condition through an image recognition technology.
Background
The stoma is caused by digestive system or urinary system diseases, and the intestinal canal needs to be separated through surgical operation treatment, and one end of the intestinal canal is led out to the body surface to form an opening. However, due to the construction of unnatural cavitary tracts, stomal complications are very common among domestic and foreign stomal patients, and the occurrence rate is as high as 68%. In daily life, an ostomy patient needs to change the ostomy bag frequently. Careful care is required at the stoma and minor negligence is very likely to cause stomal complications. The quality of life of an ostomy patient after surgery is directly affected by ostomy complications which make the patient more difficult to care and delay the time to recovery, while at the same time the patient is under a greater psychological and economic burden. Simultaneously, the patient oneself is when the pocket is made in the change of family, can't judge the condition of making a mouthful by oneself, also can't know clearly how to nurse to the different condition of making a mouthful, and then can't judge what kind of condition needs to seek medical advice to the condition such as the stoma condition deterioration, the misty treatment opportunity appears easily, if frequently go to hospital and seek medical advice, then can bring more trouble, increase economic burden for the patient, also increased the work load of making a mouthful nurse practitioner moreover. AI medical treatment is now widely used in medical image recognition, disease diagnosis, and the like at home and abroad. The AI can identify abnormal regions in the medical image, thereby providing reference for clinicians and improving the lesion detection rate. If the patient can automatically identify the stoma condition through an intelligent technology at home, the problem can be well solved, and the application of the stoma condition can be automatically identified is not perfect at present.
Disclosure of Invention
The present invention is directed to an application system for identifying a stoma situation by image recognition technology, so as to solve the problems mentioned in the background art.
To achieve the above object, one of the objects of the present invention is to provide an application system for identifying stoma by image recognition technology, comprising
The system comprises a system base unit, an identification management unit and a function application unit; the system base unit, the identification management unit and the function application unit are sequentially connected through Ethernet communication; the system basic unit is used for providing various equipment devices, technologies, applications and the like supporting system operation; the identification management unit is used for acquiring image information of the stoma and carrying out a series of processing operations on the image to obtain an identification result; the function application unit is used for adding various extended services on the basis of image recognition so as to improve the functions of the system;
the system base unit comprises a capital construction management module, a cloud database module, a training learning module and a network communication module;
the identification management unit comprises an image acquisition module, an image processing module, a comparison analysis module and a result output module;
the functional application unit comprises a quality evaluation module, a nursing guidance module, an identification statistical module and an online hospitalizing module.
As a further improvement of the technical scheme, the infrastructure management module, the cloud database module and the training learning module are sequentially connected through ethernet communication; the infrastructure management module is used for providing equipment such as a processing device and the like to serve as a system carrier and providing an intelligent technology to support smooth operation of the system; the cloud terminal database module is used for acquiring a large amount of public information related to the stoma, and forming a basic database at the cloud terminal after the information is collated; the training learning module is used for building a training model based on a deep learning algorithm and a prediction algorithm and randomly extracting a plurality of data from a database for training and deep learning; the network communication module is used for establishing a channel for connecting communication and data transmission among all layers of the system.
The equipment in the infrastructure comprises but is not limited to a processor, a display terminal, a camera, a data acquisition device, a mobile terminal with a camera shooting function and the like; techniques in the infrastructure include, but are not limited to, image recognition technology, wireless transmission technology, the Fast-RCNN algorithm, Resnet-50, RPN, Fast R-CNN, ImageNet, and the like.
The information related to the stoma includes, but is not limited to, cases of ostomy patients, ostomy pictures, ostomy disorders, ostomy abnormalities, and treatment regimens corresponding thereto.
The network communication technology includes, but is not limited to, wired communication, wireless WiFi, data traffic, bluetooth, and the like.
As a further improvement of the technical scheme, the cloud database module comprises an acquisition updating module, a cleaning and screening module, a classification and induction module and a data storage module; the signal output end of the collection updating module is connected with the signal input end of the cleaning and screening module, the signal output end of the cleaning and screening module is connected with the signal input end of the classification and summary module, and the signal output end of the classification and summary module is connected with the signal input end of the data storage module; the acquisition updating module is used for acquiring mass data from medical center official networks and networks to establish a database and updating the newly added data of each source into the database; the cleaning and screening module is used for cleaning the data in the database at regular time to screen out repeated, invalid, overdue, missing and wrong data; the classification induction module is used for classifying and inducing the standards set by the data; the data storage module is used for respectively creating a plurality of folders according to set standards and storing data into the corresponding folders.
As a further improvement of the technical solution, the cleaning and screening module adopts an entropy algorithm of information quantity, and a calculation formula thereof is as follows:
H(x)＝-∑P(Xi)log2P(Xi)；
wherein, i is 1,2,3iDenotes the ith state (n states in total), P (X)i) Represents the probability of the i-th state occurring, and h (x) is the amount of information needed to remove uncertainty, in bits (bits).
As a further improvement of the technical solution, the classification induction module adopts an ID3 algorithm, and the algorithm flow is as follows:
let S be a set of S data samples, defining m different classes Ci(i ═ 1,2,. multidot.m), let siIs CiThe number of samples in a class, then the desired information value for a given sample S is calculated by:
wherein p isiIs that any sample belongs to CiProbability of pi＝si/s；
Let attribute A have different values { a }1，a2,., a }, the sample S may be divided into { S with attribute A1，S2，...，SVIs given by sijIs SjC iniThe number of samples of the class, the entropy divided into subsets by a is calculated by:
as a further improvement of the technical solution, a signal output end of the image acquisition module is connected with a signal input end of the graphics processing module, a signal output end of the graphics processing module is connected with a signal input end of the contrast analysis module, and a signal output end of the contrast analysis module is connected with a signal input end of the result output module; the image acquisition module is used for acquiring image information or video information of the stoma condition and intercepting clear picture information from the video information; the image processing module is used for performing operations such as cutting, gray level adjustment, characteristic delineation and the like on the acquired image so as to perform subsequent identification operation; the comparison analysis module is used for comparing the processed graph with a large amount of graph information stored in a database and achieving the purpose of identifying the type of the image through comprehensive analysis; and the result output module is used for outputting the result information obtained by analysis and feeding back the result information to the user through the display terminal.
The mode of acquiring the image can be shooting directly through a camera of the device or shooting through a mobile terminal with a camera shooting function and transmitting the shot image to the system; the photographed stoma situation image information may be a picture or a video.
As a further improvement of the technical scheme, the image processing module comprises a cutting enhancement module, a color filtering module, a feature extraction module and a sketching marking module; the signal output end of the cutting enhancement module is connected with the signal input end of the color filtering module, the signal output end of the color filtering module is connected with the signal input end of the characteristic extraction module, and the signal output end of the characteristic extraction module is connected with the signal input end of the delineation marking module; the cropping enhancement module is used for cropping and zooming the picture to enable the format of the picture to meet the standard requirement so as to enhance the contrast of the image; the color filtering module is used for filtering the colors of the image to make the colors gray so as to reduce the interference influence of the colors on the identification and judgment; the characteristic extraction module is used for extracting a characteristic region different from a surrounding organization structure in the graph; the drawing and marking module is used for drawing the characteristic region to be convex and marking the type of the characteristic possibly in the drawing region.
As a further improvement of the technical solution, the color filtering module adopts a color binarization processing method, and the processing steps are as follows:
step1, setting the gray value f (i, j) of the image at the pixel point (i, j), and considering the (2 ω +1) × (2 ω +1) window with the pixel point (i, j) as the center;
step2, calculating a threshold T (i, j) of each pixel point (i, j) in the image;
step3, each pixel point (i, j) in the image is binarized point by using the b (i, j) value.
As a further improvement of the technical scheme, the quality evaluation module, the nursing guidance module, the judgment and statistics module and the online hospitalization module are sequentially connected through ethernet communication and operate independently; the quality evaluation module is used for carrying out grading evaluation on the stoma condition and the nursing effect of the patient according to the image recognition result and carrying out symptom evaluation on the stoma with stoma complications; the nursing guidance module is used for providing nursing guidance suggestions for the patient according to the grade of the quality assessment and providing a treatment scheme for the stoma condition with complications; the judgment and statistics module is used for counting the workload of image identification of the system at regular time so as to calculate the frequency of replacing a stoma of a patient, and regularly analyzing the accuracy of the judgment work of the system; the on-line hospitalizing module is used for providing a channel for the patient to directly communicate with the ostomy care provider through a network.
Complications of the stoma include, among others, bleeding irritant dermatitis of the stoma, retraction of the stoma, prolapse of the stoma, hernia, stenosis of the stoma, necrosity of the stoma, etc.
Another object of the present invention is to provide an application method of an application system for identifying a stoma condition by an image recognition technique, comprising the steps of:
s1, acquiring massive data from each medical database platform and the network by the system, acquiring ethical approval and use permission according to Helsinki declaration, cleaning, screening, classifying and summarizing the data, storing the data into corresponding folders respectively, and forming a basic database at the cloud;
s2, building a training model in the system, randomly extracting a large amount of data from a basic database to build a training number set, introducing the training number set into the training model for training, automatically inducing rules from the data by the model, acquiring internal rules, and enabling the system to have good self-learning, self-adaption, associative memory, parallel processing and nonlinear shape conversion capabilities through deep learning;
s3, when the patient changes the ostomy bag, shooting an image or video of the ostomy condition through a camera carried by the system carrier equipment, or shooting the image or video through the mobile terminal, and transmitting the image or video on the mobile terminal to the system through a data line or a wireless transmission technology;
s4, the system receives the image or video, automatically selects a clear picture or intercepts the clear picture in the video, and sequentially performs the processing of cutting, zooming, graying, feature extraction, feature sketching, marking and the like on the picture;
s5, the system brings the processed pictures into a system which is well fitted, and the system can predict and classify the new pictures according to the learned experience and output the recognized picture type result;
s6, the system evaluates the stoma condition of the patient according to the picture recognition result and provides nursing guidance; if the patient has complications at the stoma, evaluating the symptom degree of the complications and giving a treatment scheme;
s7, if the system judgment result is the stoma complication needing medical intervention, the patient can directly communicate with the stoma care giver on the system through network connection, and during the communication, the stoma condition picture and the system identification result can be directly called and sent to the stoma care giver so that the stoma care giver can diagnose; if the on-line stomacher can not diagnose directly, the patient needs to go to the hospital for a doctor;
and S8, the system carries out statistics on the number of times of image recognition at regular time to calculate the frequency of changing the ostomy bag for the patient, analyzes according to the calculation result and gives a prompt to the patient when the result is abnormal.
It is a further object of the present invention to provide an operating device for an application system for identifying an ostomy situation by image recognition technology, comprising a processor, a memory and a computer program stored in the memory and running on the processor, wherein the processor is adapted to implement any of the above-mentioned application systems for identifying an ostomy situation by image recognition technology when executing the computer program.
It is a further object of the present invention to provide a computer-readable storage medium storing a computer program which, when executed by a processor, implements any of the above-described stoma situation determination application systems by image recognition technology.
Compared with the prior art, the invention has the beneficial effects that: in the application system for identifying the stoma condition through the image recognition technology, the patient can upload the image information of the stoma condition to the system by creating a large database of the stoma case in the system and building a training model on the basis of the large database when changing the stoma bag every time, the system can process and identify the image information and evaluate and provide nursing guidance suggestions and treatment schemes in time, and in addition, a channel for seeing a doctor on the internet can be provided for the patient, so that the patient can judge the stoma condition at home, the stoma is convenient to better nurse, the probability of stoma deterioration is reduced, the treatment opportunity is avoided being delayed, the economic burden caused by frequent doctor seeing the doctor can be reduced, the workload of a stoma caregiver is reduced, the problem can be found at the early stage of complication occurrence in the stoma, early warning is provided for the patient, and timely and effective medical professional support is obtained, the workload of clinical nursing workers can be greatly reduced, and the problems that public doctors are difficult to see and the workload of nurses in large hospitals is heavy are solved.
Drawings
FIG. 1 is an exemplary product architecture diagram of the present invention;
FIG. 2 is a block diagram of the overall system apparatus of the present invention;
FIG. 3 is a diagram of one embodiment of a local system device architecture;
FIG. 4 is a second block diagram of a local system apparatus according to the present invention;
FIG. 5 is a third block diagram of a local system apparatus according to the present invention;
FIG. 6 is a fourth embodiment of the present invention;
FIG. 7 is a fifth embodiment of the present invention;
FIG. 8 is a block diagram of an exemplary computer program product of the present invention.
The various reference numbers in the figures mean:
1. processing the host; 2. a display terminal; 3. a camera; 4. a cloud database; 5. training a model; 6. a mobile terminal;
100. a system base unit; 101. a capital construction management module; 102. a cloud database module; 1021. an acquisition updating module; 1022. cleaning and screening the module; 1023. a classification and induction module; 1024. a data storage module; 103. training a learning module; 104. a network communication module;
200. an identification management unit; 201. an image acquisition module; 202. a graphics processing module; 2021. a cutting enhancement module; 2022. a color filtering module; 2023. a feature extraction module; 2024. a delineation marking module; 203. a comparison analysis module; 204. a result output module;
300. a function application unit; 301. a quality evaluation module; 302. a care guidance module; 303. an identification statistical module; 304. and an online hospitalizing 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.
System embodiment
As shown in FIGS. 1 to 8, the present embodiment provides an application system for recognizing a stoma situation by image recognition technology, comprising
A system base unit 100, an identification management unit 200, and a function application unit 300; the system base unit 100, the identification management unit 200 and the function application unit 300 are sequentially connected through ethernet communication; the system base unit 100 is used for providing various equipment devices, technologies, applications, and the like supporting system operation; the recognition management unit 200 is for acquiring image information of the stoma and performing a series of processing operations on the image to obtain a recognition result; the function application unit 300 is used for adding a plurality of extended services on the basis of image recognition to complete the functions of the system;
the system base unit 100 comprises a capital construction management module 101, a cloud database module 102, a training learning module 103 and a network communication module 104;
the recognition management unit 200 comprises an image acquisition module 201, a graphic processing module 202, a comparison analysis module 203 and a result output module 204;
the functional application unit 300 includes a quality evaluation module 301, a care guidance module 302, an opinion statistics module 303, and an online hospitalization module 304.
In this embodiment, the infrastructure management module 101, the cloud database module 102, and the training learning module 103 are sequentially connected through ethernet communication; the infrastructure management module 101 is used for providing equipment such as a processing device and the like as a system carrier and providing an intelligent technology to support smooth operation of the system; the cloud terminal database module 102 is used for acquiring a large amount of public information related to the stoma, and forming a basic database at the cloud terminal after the information is collated; the training learning module 103 is used for building a training model based on a deep learning algorithm and a prediction algorithm and randomly extracting a plurality of data from a database for training and deep learning; the network communication module 104 is used to establish a channel for connecting communication and data transmission between the layers of the system.
Specifically, the artificial intelligence deep learning algorithm FasterRCNN is built to train the stoma image data, and the method mainly comprises the following three processes: step1, extracting image features by using a convolutional neural network, and extracting the features of the image of the marked stoma case, wherein the image comprises an artery phase, a vein phase and a delay phase; step2, introducing anchor points through RPN, mapping to original images, training through anchor point frames, performing two classifications, judging whether the contents of candidate frames are foreground or background, leaving the candidate frames of the foreground, abandoning the candidate frames of the background, and finely tuning BBox of the foreground to be close to the marked real frames through regression; step3 finds the specific location of the detection box by ROI pooling, classification and regression.
Wherein, the ROI pooling is mainly used for extracting representative features of the target object.
And then, on one hand, calculating the probability that each candidate region belongs to each category through the full connection layer and softmax, outputting the classification category, and on the other hand, obtaining the position information of the candidate frame by utilizing regression, so as to obtain the final accurate position of the detection frame.
Further, the framework of the whole model applies Resnet-50 to perform feature extraction, RPN generation feature region proposal and Fast R-CNN network completion bounding box classification and regression for the main network framework. The network has completed pre-training in the ImageNet image database. And (3) incorporating the training set data into the model for training, adjusting model parameters such as the number of neuron layers, the size of a convolution kernel, a loss function value and the like for multiple times to make the model reach convergence in each classification group, and establishing the stoma identification system.
And finally, evaluating the robustness and the prediction efficiency of the model by using the test set data, and taking the area under the working characteristic curve (AUC), the Accuracy (Accuracy), the Precision (Precision), the Recall (Recall) and the F1-index of the subject as quantitative model evaluation indexes.
The equipment in the infrastructure comprises but is not limited to a processor, a display terminal, a camera, a data acquisition device, a mobile terminal with a camera shooting function and the like; techniques in the infrastructure include, but are not limited to, image recognition technology, wireless transmission technology, the Fast-RCNN algorithm, Resnet-50, RPN, Fast R-CNN, ImageNet, and the like.
The information related to the stoma includes, but is not limited to, cases of ostomy patients, ostomy pictures, ostomy disorders, ostomy abnormalities, and treatment regimens corresponding thereto.
The network communication technology includes, but is not limited to, wired communication, wireless WiFi, data traffic, bluetooth, and the like.
Further, the cloud database module 102 includes an acquisition update module 1021, a cleaning and screening module 1022, a classification induction module 1023, and a data storage module 1024; the signal output end of the acquisition and update module 1021 is connected with the signal input end of the cleaning and screening module 1022, the signal output end of the cleaning and screening module 1022 is connected with the signal input end of the classification and induction module 1023, and the signal output end of the classification and induction module 1023 is connected with the signal input end of the data storage module 1024; the acquisition and update module 1021 is used for acquiring mass data from each medical center official website and the internet to establish a database and updating the newly added data of each source into the database; the cleaning and screening module 1022 is used for cleaning the data in the database at regular time to screen out the repeated, invalid, overdue, missing and wrong data; the classification induction module 1023 is used for classifying and inducing the standard set by the data; the data storage module 1024 is configured to create a plurality of folders according to a set standard and store data in the corresponding folders.
Specifically, the cleaning and screening module 1022 uses an entropy algorithm of information quantity, and the calculation formula is as follows:
H(x)＝-∑P(Xi)log2P(Xi)；
wherein, i is 1,2,3iDenotes the ith state (n states in total), P (X)i) Represents the probability of the i-th state occurring, and h (x) is the amount of information needed to remove uncertainty, in bits (bits).
Specifically, the classification induction module 1023 adopts an ID3 algorithm, and the algorithm flow is as follows:
let S be a set of S data samples, defining m different classes Ci(i ═ 1,2,. multidot.m), let siIs CiThe number of samples in a class, then the desired information value for a given sample S is calculated by:
wherein p isiIs that any sample belongs to CiProbability of pi＝si/s；
Let attribute A have different values { a }1，a2,., a }, the sample S may be divided into { S with attribute A1，S2，...，SVIs given by sijIs SjC iniThe number of samples of the class, the entropy divided into subsets by a is calculated by:
in this embodiment, the signal output end of the image acquisition module 201 is connected to the signal input end of the graphics processing module 202, the signal output end of the graphics processing module 202 is connected to the signal input end of the contrast analysis module 203, and the signal output end of the contrast analysis module 203 is connected to the signal input end of the result output module 204; the image acquisition module 201 is used for acquiring image information or video information of a stoma condition and intercepting clear picture information from the video information; the image processing module 202 is configured to perform operations such as clipping, gray scale adjustment, feature delineation, and the like on the acquired image so as to perform subsequent recognition operations; the comparison analysis module 203 is used for comparing the processed graph with a large amount of graph information stored in a database and achieving the purpose of identifying the type of the image through comprehensive analysis; the result output module 204 is configured to output the result information obtained by the analysis and feed the result information back to the user through the display terminal.
The mode of acquiring the image can be shooting directly through a camera of the device or shooting through a mobile terminal with a camera shooting function and transmitting the shot image to the system; the photographed stoma situation image information may be a picture or a video.
Further, the graphics processing module 202 includes a cropping enhancement module 2021, a color filtering module 2022, a feature extraction module 2023, and a sketching labeling module 2024; the signal output end of the cutting enhancement module 2021 is connected with the signal input end of the color filtering module 2022, the signal output end of the color filtering module 2022 is connected with the signal input end of the feature extraction module 2023, and the signal output end of the feature extraction module 2023 is connected with the signal input end of the delineation marking module 2024; the cropping enhancement module 2021 is used for cropping and scaling the picture to make the format meet the standard requirement so as to enhance the contrast of the image; the color filtering module 2022 is used for filtering the colors of the image to make the colors grayed so as to reduce the interference influence of the colors on the identification judgment; the feature extraction module 2023 is configured to extract a feature region in the graph that is different from a surrounding tissue structure; the delineation marking module 2024 is used to delineate the region of the feature so that it is salient and mark the type of the feature in the delineated region that may belong to.
Specifically, the specific operation steps of the graphic processing are as follows: drawing the stoma part (region of interest) of each image by using LabelIMG software, wherein the drawing work is completed by two senior nurse technicians together, the drawing mode is rectangular frame drawing, the edge of the stoma does not need to be finely traced, the information drawn by each image is stored in a file form of xml format, and the coordinate information drawn and the classified label information are recorded in each xml file; dividing the image data into a training set and a test set according to the ratio of 4: 1; the image is reduced to be a square with 512 x 512 pixels by using bilinear interpolation (bilinear interpolation), data enhancement is carried out by horizontal turning, vertical turning, shearing and scaling conversion, the random rotation angle of the image is 40 degrees, and the amplitude of horizontal, vertical shifting, shearing conversion and random scaling is 0.2, so that the size of the data set is expanded, and the generalization capability of the model is enhanced.
Specifically, the color filtering module 2022 adopts a color binarization processing method, and the processing steps are as follows:
step1, setting the gray value f (i, j) of the image at the pixel point (i, j), and considering the (2 ω +1) × (2 ω +1) window with the pixel point (i, j) as the center;
step2, calculating a threshold T (i, j) of each pixel point (i, j) in the image;
step3, each pixel point (i, j) in the image is binarized point by using the b (i, j) value.
In this embodiment, the quality evaluation module 301, the nursing guidance module 302, the judgment and statistics module 303 and the online hospitalization module 304 are sequentially connected through ethernet communication and operate independently; the quality evaluation module 301 is used for performing grading evaluation on the stoma condition and the nursing effect of the patient according to the image recognition result, and performing symptom evaluation on the stoma with stoma complications; the nursing guidance module 302 is used for providing nursing guidance suggestions for the patients according to the grade of the quality assessment and providing treatment schemes for the stoma condition with complications; the judgment and statistics module 303 is used for counting the workload of image identification performed by the system at regular time to calculate the frequency of replacing the stoma of the patient, and regularly analyzing the accuracy of the judgment work of the system; the online hospitalization module 304 is used to provide the patient with a channel for direct communication with the ostomy care giver via the network.
Among the common stomal complications are: stomal skin lesions, stomal hemorrhage, stomal ischemic necrosis, stomal skin mucosa separation, stomal periphery dermatitis, stomal retraction, parastomal hernia, stomal prolapse, stomal stenosis, etc.; and among the highest incidence are peri-stomal skin complications, including: fecal dermatitis, folliculitis, fungal infection, allergic dermatitis, mechanical injury, etc.
Method embodiment
The present embodiment aims to provide an application method of an application system for identifying a stoma situation through an image recognition technology, which comprises the following steps:
s1, acquiring massive data from each medical database platform and the network by the system, acquiring ethical approval and use permission according to Helsinki declaration, cleaning, screening, classifying and summarizing the data, storing the data into corresponding folders respectively, and forming a basic database at the cloud;
s2, building a training model in the system, randomly extracting a large amount of data from a basic database to build a training number set, introducing the training number set into the training model for training, automatically inducing rules from the data by the model, acquiring internal rules, and enabling the system to have good self-learning, self-adaption, associative memory, parallel processing and nonlinear shape conversion capabilities through deep learning;
s3, when the patient changes the ostomy bag, shooting an image or video of the ostomy condition through a camera carried by the system carrier equipment, or shooting the image or video through the mobile terminal, and transmitting the image or video on the mobile terminal to the system through a data line or a wireless transmission technology;
s4, the system receives the image or video, automatically selects a clear picture or intercepts the clear picture in the video, and sequentially performs the processing of cutting, zooming, graying, feature extraction, feature sketching, marking and the like on the picture;
s5, the system brings the processed pictures into a system which is well fitted, and the system can predict and classify the new pictures according to the learned experience and output the recognized picture type result;
s6, the system evaluates the stoma condition of the patient according to the picture recognition result and provides nursing guidance; if the patient has complications at the stoma, evaluating the symptom degree of the complications and giving a treatment scheme;
s7, if the system judgment result is the stoma complication needing medical intervention, the patient can directly communicate with the stoma care giver on the system through network connection, and during the communication, the stoma condition picture and the system identification result can be directly called and sent to the stoma care giver so that the stoma care giver can diagnose; if the on-line stomacher can not diagnose directly, the patient needs to go to the hospital for a doctor;
and S8, the system carries out statistics on the number of times of image recognition at regular time to calculate the frequency of changing the ostomy bag for the patient, analyzes according to the calculation result and gives a prompt to the patient when the result is abnormal.
Computer program product embodiment
Referring to fig. 1, an exemplary product architecture diagram of an application system for identifying a stoma condition through image recognition technology is shown, the system comprises a processing host 1 and a display terminal 2 matched with the processing host, the display terminal 2 is provided with a camera 3, the camera 3 can be external or installed in the display terminal 2, the processing host 1 is connected with a cloud database 4 through ethernet communication, the processing host 1 is also provided with a training model 5 built based on the cloud database 4, the processing host 1 is also provided with a mobile terminal 6, and the mobile terminal 6 can be connected with the processing host 1 through data lines or wireless communication.
Referring to fig. 8, there is shown a schematic diagram of an operating apparatus of an application system for identifying a stoma situation by means of image recognition technology, the apparatus comprising a processor, a memory and a computer program stored in the memory and running on the processor.
The processor comprises one or more processing cores, the processor is connected with the processor through a bus, the memory is used for storing program instructions, and the application system for identifying the stoma condition through the image recognition technology is realized when the processor executes the program instructions in the memory.
Alternatively, the memory may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
Furthermore, the invention provides a computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, implements a system for applying identification of a stoma situation by means of image recognition techniques as described above.
Alternatively, the present invention also provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the above-described aspects of the application system for identifying an ostomy situation by image recognition techniques.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, and the program may be stored in a computer-readable storage medium, where the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and the preferred embodiments of the present invention are described in the above embodiments and the description, and are not intended to limit the present invention. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. An application system for identifying a stoma situation by means of image recognition technology, characterized in that: comprises that
A system base unit (100), an identification management unit (200) and a function application unit (300); the system base unit (100), the identification management unit (200) and the function application unit (300) are sequentially connected through Ethernet communication; the system basic unit (100) is used for providing various equipment devices, technologies, applications and the like supporting system operation; the identification management unit (200) is used for acquiring image information of the stoma and carrying out a series of processing operations on the image to obtain an identification result; the function application unit (300) is used for adding a plurality of extended services on the basis of image recognition to complete the functions of the system;
the system base unit (100) comprises a capital construction management module (101), a cloud database module (102), a training learning module (103) and a network communication module (104);
the recognition management unit (200) comprises an image acquisition module (201), a graphic processing module (202), a comparison analysis module (203) and a result output module (204);
the function application unit (300) comprises a quality evaluation module (301), a nursing guidance module (302), an identification statistic module (303) and an online hospitalization module (304).
2. The stoma circumstance application system according to claim 1, wherein: the infrastructure management module (101), the cloud database module (102) and the training learning module (103) are sequentially connected through Ethernet communication; the infrastructure management module (101) is used for providing equipment such as a processing device and the like to serve as a system carrier and providing an intelligent technology to support smooth operation of the system; the cloud database module (102) is used for acquiring a large amount of public stoma related information and forming a basic database in the cloud after the information is collated; the training learning module (103) is used for building a training model based on a deep learning algorithm and a prediction algorithm and randomly extracting a plurality of data from a database for training and deep learning; the network communication module (104) is used for establishing a channel for connecting communication and data transmission among all layers of the system.
3. The stoma circumstance application system according to claim 2, wherein: the cloud database module (102) comprises a collection updating module (1021), a cleaning and screening module (1022), a classification induction module (1023) and a data storage module (1024); the signal output end of the collection updating module (1021) is connected with the signal input end of the cleaning and screening module (1022), the signal output end of the cleaning and screening module (1022) is connected with the signal input end of the classification and induction module (1023), and the signal output end of the classification and induction module (1023) is connected with the signal input end of the data storage module (1024); the acquisition updating module (1021) is used for acquiring mass data from official networks and networks of various medical centers to establish a database and updating newly added data of various sources into the database; the cleaning and screening module (1022) is used for cleaning the data in the database at regular time to screen out repeated, invalid, overdue, missing and wrong data; the classification induction module (1023) is used for classifying and inducing the standard set by the data; the data storage module (1024) is used for respectively creating a plurality of folders according to set standards and storing data into the corresponding folders.
4. The stoma circumstance application system according to claim 3, wherein: the cleaning and screening module (1022) adopts an entropy algorithm of information quantity, and the calculation formula is as follows:
wherein, i is 1,2,3iDenotes the ith state (n states in total), P (X)i) Represents the probability of the i-th state occurring, and h (x) is the amount of information needed to remove uncertainty, in bits (bits).
5. The stoma circumstance application system according to claim 3, wherein: the classification induction module (1023) adopts an ID3 algorithm, and the algorithm flow is as follows:
let S be a set of S data samples, defining m different classes Ci(i ═ 1,2,. multidot.m), let siIs CiThe number of samples in a class, then the desired information value for a given sample S is calculated by:
wherein p isiIs that any sample belongs to CiProbability of pi＝si/s；
Let attribute A have different values { a }1，a2,., a }, the sample S may be divided into { S with attribute A1，S2，...，SVIs given by sijIs SjC iniThe number of samples of the class, the entropy divided into subsets by a is calculated by:
6. the stoma circumstance application system according to claim 1, wherein: the signal output end of the image acquisition module (201) is connected with the signal input end of the graphics processing module (202), the signal output end of the graphics processing module (202) is connected with the signal input end of the comparison analysis module (203), and the signal output end of the comparison analysis module (203) is connected with the signal input end of the result output module (204); the image acquisition module (201) is used for acquiring image information or video information of a stoma condition and intercepting clear picture information from the video information; the image processing module (202) is used for performing operations such as cutting, gray level adjustment and feature delineation on the acquired image so as to perform subsequent identification operation; the comparison analysis module (203) is used for comparing the processed graph with a large amount of graph information stored in a database and achieving the purpose of identifying the type of the image through comprehensive analysis; and the result output module (204) is used for outputting the result information obtained by analysis and feeding back the result information to the user through the display terminal.
7. The stoma circumstance application system according to claim 6, wherein: the graphics processing module (202) comprises a cropping enhancement module (2021), a color filtering module (2022), a feature extraction module (2023) and a sketching marking module (2024); the signal output end of the cutting enhancement module (2021) is connected with the signal input end of the color filtering module (2022), the signal output end of the color filtering module (2022) is connected with the signal input end of the feature extraction module (2023), and the signal output end of the feature extraction module (2023) is connected with the signal input end of the delineation marking module (2024); the cropping enhancement module (2021) is used for cropping and scaling the picture to enable the format of the picture to meet the standard requirement so as to enhance the contrast of the image; the color filtering module (2022) is used for filtering the colors of the image to graying so as to reduce the interference influence of the colors on the identification judgment; the feature extraction module (2023) is used for extracting a feature region different from a surrounding tissue structure in the graph; the delineation marking module (2024) is used for delineating the characteristic region to be convex and marking the type of the characteristic possibly belonging to the delineation region.
8. The stoma circumstance application system according to claim 7, wherein: the color filtering module (2022) adopts a color binarization processing method, and the processing steps are as follows:
step1, setting the gray value f (i, j) of the image at the pixel point (i, j), and considering the (2 ω +1) × (2 ω +1) window with the pixel point (i, j) as the center;
step2, calculating a threshold T (i, j) of each pixel point (i, j) in the image;
step3, each pixel point (i, j) in the image is binarized point by using the b (i, j) value.
9. The stoma circumstance application system according to claim 1, wherein: the quality evaluation module (301), the nursing guidance module (302), the judgment statistics module (303) and the online hospitalization module (304) are sequentially connected through Ethernet communication and operate independently; the quality evaluation module (301) is used for carrying out grading evaluation on the stoma condition and the nursing effect of the patient according to the image recognition result and carrying out symptom evaluation on the stoma with stoma complications; the nursing guidance module (302) is used for providing nursing guidance for the patient according to the grade of the quality assessment and providing a treatment scheme for the stoma condition with complications; the judgment and statistics module (303) is used for counting the workload of image identification of the system at regular time to calculate the frequency of the patient for replacing the stoma, and regularly counting the accuracy of the judgment work of the system; the online hospitalization module (304) is for providing a patient with a channel for direct communication with an ostomy care provider via a network.
10. The stoma circumstance application system according to claim 1, wherein: the application method of the system comprises the following steps:
s1, acquiring massive data from each medical database platform and the network by the system, acquiring ethical approval and use permission according to Helsinki declaration, cleaning, screening, classifying and summarizing the data, storing the data into corresponding folders respectively, and forming a basic database at the cloud;
s2, building a training model in the system, randomly extracting a large amount of data from a basic database to build a training number set, introducing the training number set into the training model for training, automatically inducing rules from the data by the model, acquiring internal rules, and enabling the system to have good self-learning, self-adaption, associative memory, parallel processing and nonlinear shape conversion capabilities through deep learning;
s3, when the patient changes the ostomy bag, shooting an image or video of the ostomy condition through a camera carried by the system carrier equipment, or shooting the image or video through the mobile terminal, and transmitting the image or video on the mobile terminal to the system through a data line or a wireless transmission technology;
s4, the system receives the image or video, automatically selects a clear picture or intercepts the clear picture in the video, and sequentially performs the processing of cutting, zooming, graying, feature extraction, feature sketching, marking and the like on the picture;
s5, the system brings the processed pictures into a system which is well fitted, and the system can predict and classify the new pictures according to the learned experience and output the recognized picture type result;
s6, the system evaluates the stoma condition of the patient according to the picture recognition result and provides nursing guidance; if the patient has complications at the stoma, evaluating the symptom degree of the complications and giving a treatment scheme;
s7, if the system judgment result is the stoma complication needing medical intervention, the patient can directly communicate with the stoma care giver on the system through network connection, and during the communication, the stoma condition picture and the system identification result can be directly called and sent to the stoma care giver so that the stoma care giver can diagnose; if the on-line stomacher can not diagnose directly, the patient needs to go to the hospital for a doctor;
and S8, the system carries out statistics on the number of times of image recognition at regular time to calculate the frequency of changing the ostomy bag for the patient, analyzes according to the calculation result and gives a prompt to the patient when the result is abnormal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110524013.9A CN113326745A (en) | 2021-05-13 | 2021-05-13 | Application system for judging and identifying stoma situation through image identification technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110524013.9A CN113326745A (en) | 2021-05-13 | 2021-05-13 | Application system for judging and identifying stoma situation through image identification technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113326745A true CN113326745A (en) | 2021-08-31 |
Family
ID=77415554
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110524013.9A Pending CN113326745A (en) | 2021-05-13 | 2021-05-13 | Application system for judging and identifying stoma situation through image identification technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113326745A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113962992A (en) * | 2021-12-21 | 2022-01-21 | 青岛大学附属医院 | Urinary calculus flat scanning CT image recognition system based on deep learning and training method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109637644A (en) * | 2018-12-12 | 2019-04-16 | 曾鹏 | A kind of cerebral apoplexy data management system and management method based on mobile Internet |
| CN109686438A (en) * | 2019-03-01 | 2019-04-26 | 徐文 | A kind of colostomy patient management system based on cloud and wearable technology |
| US20190240059A1 (en) * | 2018-02-02 | 2019-08-08 | 11 Health and Technologies Inc. | Ostomy patient care system and method |
| CN110556031A (en) * | 2019-09-06 | 2019-12-10 | 重庆大学附属肿瘤医院 | Medical guidance system for ostomy patient |
| CN111914274A (en) * | 2020-07-30 | 2020-11-10 | 南京中诚区块链研究院有限公司 | Full-process block chain system based on multiple information sources |
| CN112102954A (en) * | 2020-09-02 | 2020-12-18 | 南京江北新区科技投资集团有限公司 | Big data analysis cloud platform system capable of providing intelligent medical service |
-
2021
- 2021-05-13 CN CN202110524013.9A patent/CN113326745A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190240059A1 (en) * | 2018-02-02 | 2019-08-08 | 11 Health and Technologies Inc. | Ostomy patient care system and method |
| CN109637644A (en) * | 2018-12-12 | 2019-04-16 | 曾鹏 | A kind of cerebral apoplexy data management system and management method based on mobile Internet |
| CN109686438A (en) * | 2019-03-01 | 2019-04-26 | 徐文 | A kind of colostomy patient management system based on cloud and wearable technology |
| CN110556031A (en) * | 2019-09-06 | 2019-12-10 | 重庆大学附属肿瘤医院 | Medical guidance system for ostomy patient |
| CN111914274A (en) * | 2020-07-30 | 2020-11-10 | 南京中诚区块链研究院有限公司 | Full-process block chain system based on multiple information sources |
| CN112102954A (en) * | 2020-09-02 | 2020-12-18 | 南京江北新区科技投资集团有限公司 | Big data analysis cloud platform system capable of providing intelligent medical service |
Non-Patent Citations (3)
| Title |
|---|
| 姚树春等: "《大数据技术与应用》", 30 June 2018, pages: 70 - 71 * |
| 孙建国: "《数字化智能放疗》", 28 February 2020, pages: 241 * |
| 肖洁银: "造口周围皮肤标准护理数字化的临床探讨", 《中国现代医生》, 31 May 2016 (2016-05-31) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113962992A (en) * | 2021-12-21 | 2022-01-21 | 青岛大学附属医院 | Urinary calculus flat scanning CT image recognition system based on deep learning and training method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10722181B2 (en) | Systems, methods, and computer-readable media for using descriptors to identify when a subject is likely to have a dysmorphic feature | |
| US11468693B2 (en) | Digital image classification method for cervical fluid-based cells based on a deep learning detection model | |
| WO2020253629A1 (en) | Detection model training method and apparatus, computer device, and storage medium | |
| CN111985536B (en) | Based on weak supervised learning gastroscopic pathology image Classification method | |
| CN107886503A (en) | A kind of alimentary canal anatomical position recognition methods and device | |
| US20220198230A1 (en) | Auxiliary detection method and image recognition method for rib fractures based on deep learning | |
| CN111429407B (en) | Chest X-ray disease detection device and method based on double-channel separation network | |
| CN107330449A (en) | A kind of BDR sign detection method and device | |
| CN109859168A (en) | A kind of X-ray rabat picture quality determines method and device | |
| US20220172828A1 (en) | Endoscopic image display method, apparatus, computer device, and storage medium | |
| CN104299242B (en) | Fluoroscopic visualization eye fundus image extracting method based on NGC ACM | |
| CN112365471B (en) | Cervical cancer cell intelligent detection method based on deep learning | |
| US20240005494A1 (en) | Methods and systems for image quality assessment | |
| CN112381762A (en) | CT rib fracture auxiliary diagnosis system based on deep learning algorithm | |
| CN112102332A (en) | Cancer WSI segmentation method based on local classification neural network | |
| CN110796659A (en) | Method, device, equipment and storage medium for identifying target detection result | |
| CN113326745A (en) | Application system for judging and identifying stoma situation through image identification technology | |
| CN113435469A (en) | Kidney tumor enhanced CT image automatic identification system based on deep learning and training method thereof | |
| CN113554641B (en) | Pediatric pharyngeal image acquisition method and device | |
| Biswas et al. | Wound area segmentation using 4-d probability map and superpixel region growing | |
| CN112330629A (en) | Facial nerve disease rehabilitation condition static detection system based on computer vision | |
| Gunasekara et al. | A feasibility study for deep learning based automated brain tumor segmentation using magnetic resonance images | |
| CN112862787B (en) | CTA image data processing method, device and storage medium | |
| CN117974603A (en) | Digestive tract multiple disease detection and segmentation method, device, equipment and storage medium | |
| CN116665881A (en) | Deep learning-based infant hemangioma skin damage picture analysis method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210831 |
|
| RJ01 | Rejection of invention patent application after publication |