CN111370098A - Pathological diagnosis system and method based on edge side calculation and service device - Google Patents
Pathological diagnosis system and method based on edge side calculation and service device Download PDFInfo
- Publication number
- CN111370098A CN111370098A CN202010344137.4A CN202010344137A CN111370098A CN 111370098 A CN111370098 A CN 111370098A CN 202010344137 A CN202010344137 A CN 202010344137A CN 111370098 A CN111370098 A CN 111370098A
- Authority
- CN
- China
- Prior art keywords
- edge side
- edge
- diagnosis
- pathological section
- computing
- 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
- 238000004364 calculation method Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000010827 pathological analysis Methods 0.000 title claims abstract description 16
- 238000003745 diagnosis Methods 0.000 claims abstract description 92
- 230000001575 pathological effect Effects 0.000 claims abstract description 80
- 238000012549 training Methods 0.000 claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 20
- 238000012360 testing method Methods 0.000 claims description 43
- 230000007170 pathology Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000012937 correction Methods 0.000 abstract description 2
- 230000010354 integration Effects 0.000 abstract 2
- 238000012216 screening Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000003902 lesion Effects 0.000 description 2
- 230000002194 synthesizing effect Effects 0.000 description 2
- 206010028980 Neoplasm Diseases 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010562 histological examination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- 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
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
-
- 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
- G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
-
- 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
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- 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
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/60—ICT specially adapted for the handling or processing of medical references relating to pathologies
-
- 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
- G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Biomedical Technology (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Pathology (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioethics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The application discloses a pathological diagnosis system and method based on an edge side calculation and service device, wherein the system comprises: the system comprises a digital slice scanner, an edge side computing terminal, a doctor diagnosis workstation and an edge side integration server; the digital slice scanner is used for scanning pathological sections into pathological section digital images; the edge side computing terminal is used for carrying out edge computing processing on the pathological section digital image and giving a preliminary diagnosis result; the doctor diagnosis workstation is used for carrying out medical diagnosis by a doctor according to the digital image of the pathological section after edge calculation; and the edge side integration server is used for training and updating the model of the edge computing terminal. The invention combines the edge calculation technology with the digital section scanner, realizes focus identification and preliminary screening diagnosis during scanning and storage, can be butted with a doctor diagnosis platform and a remote consultation system, and greatly improves the pathological section identification efficiency. And a joint learning mechanism is adopted, the training of the plurality of hospital workstation models is carried out simultaneously, and corresponding correction parameters are returned. The system can obtain a model with high robustness and high identification precision.
Description
Technical Field
The invention belongs to the crossing field of medical image identification, edge calculation and automatic system engineering, and particularly relates to a pathological diagnosis system and method based on an edge side calculation and service device.
Background
Pathological section examination has a very wide application in medicine, and is very important in histological examination of diseases and judgment of tumor properties. However, the traditional pathological section is not easy to store, abrasion is caused by multiple use, and a user can not observe a plurality of people simultaneously by using a microscope when observing the pathological section. These disadvantages are inevitable, and with the development of society, the pathological section can avoid the disadvantages after being digitalized, and the pathological section can facilitate the future treatment and research of researchers.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a pathological diagnosis system and method based on an edge side computing device. Wherein, a pathological diagnosis system based on the edge side calculation device comprises: the system comprises a digital slice scanner, an edge side computing terminal, a doctor diagnosis workstation and an edge side integrated computing server;
the digital slice scanner is connected with the edge side computing terminal, and the edge side computing terminal is respectively connected with the doctor diagnosis workstation and the edge side integrated server;
the digital slice scanner is used for scanning pathological sections into pathological section digital images and transmitting the pathological section digital images to the edge side computing terminal;
the edge side computing terminal receives the pathological section digital image transmitted by the digital section scanner, carries out edge computing processing on the pathological section digital image, gives a preliminary diagnosis result comprising a recognition result and a segmentation result of a focus part, and transmits the pathological section digital image and the preliminary diagnosis result after the edge computing processing to a doctor diagnosis workstation; meanwhile, receiving a doctor diagnosis result fed back by the doctor diagnosis workstation, confirming the marked part of the doctor diagnosis result, synthesizing the preliminary diagnosis result and the doctor diagnosis result into a test result, encrypting the test result and transmitting the encrypted test result to the edge side integrated computing server;
and in the doctor diagnosis workstation, a doctor carries out medical diagnosis according to the digital image of the pathological section after edge calculation and transmits the diagnosis result of the doctor to the edge side calculation terminal.
The edge side integrated computing server: and receiving the encrypted test result transmitted by the edge side computing terminal, performing a new round of training on the basis of the previously trained model, and transmitting the model parameters after the new round of training to the edge side computing terminal.
The edge side integrated computing server is placed at a far end;
the edge side computing terminal is placed inside the hospital.
The edge side integrated computing server is characterized in that multiple hospitals adopt the same edge integrated computing server, encrypted test results of all the hospitals are transmitted to the same edge integrated computing server, the edge side integrated computing server conducts a new round of training again by using the encrypted test results of all the hospitals, and model parameters after the new round of training are transmitted to edge side computing terminals of all the hospitals.
The edge side integrated computing server filters patient information when a new round of training is performed, and transmits the patient information to the edge side computing device in each hospital system through the secure link, so that the purpose of protecting the privacy of patients is achieved.
The secure link, comprising:
(1) filtering out basic information of the patient, such as: patient name, age, gender, etc.
(2) Carrying out encryption operation on information in the secure link, wherein the secure link can be opened only through a secret key;
firstly, carrying out hash operation on an original file; secondly, generating a random number to encrypt the original file; and finally, encrypting the random number by using the public key to ensure that only the edge side integrated computing server can open the secure link.
A pathological diagnosis method based on an edge side computing device is realized by adopting the pathological diagnosis system based on the edge side computing device, and comprises the following specific steps:
step S1: scanning the pathological section into a pathological section digital image by using a digital section scanner, and transmitting the pathological section digital image to the edge side computing terminal;
step S2: using an edge side computing terminal to receive the digital image of the pathological section transmitted by the digital section scanner, carrying out edge computing processing on the digital image of the pathological section, giving a preliminary diagnosis result comprising a lesion part identification result and a segmentation result, and transmitting the digital image of the pathological section and the preliminary diagnosis result after the edge computing processing to a doctor diagnosis workstation;
step S3: through the doctor diagnosis workstation, a doctor carries out medical diagnosis according to the digital image of the pathological section after the edge calculation, and transmits the diagnosis result of the doctor to the edge side calculation terminal.
Step S4: and (4) integrating the doctor diagnosis result fed back by the doctor diagnosis workstation and the preliminary diagnosis result generated in the step (S2) by using the edge side computing terminal to generate a test result, encrypting the test result and transmitting the encrypted test result to the edge side integrated computing server.
Step S5: and carrying out a new round of model training by using the edge side integrated computing server according to the encrypted test result transmitted by the edge side computing terminal, and transmitting the model parameters after the new round of training to the edge side computing terminal.
In step S2, the process of performing edge calculation processing and giving a preliminary diagnosis result is as follows:
step S2.1: carrying out standardized preprocessing on the digital image of the pathological section;
step S2.1.1: normalizing the digital image of the pathological section;
step S2.1.2: sharpening the normalized digital image of the pathological section;
step S2.1.3: and cutting the sharpened digital image into a specified size.
Step S2.2: performing edge calculation processing on the preprocessed digital image of the pathological section, and giving a preliminary diagnosis result;
s2.2.1: setting initial model parameters;
s2.2.2: placing the preprocessed pathological section digital images into a U-net network for testing, and outputting a preliminary diagnosis result, namely a plurality of images with pathological part identification and segmentation; the U-net network is an initial network model and model parameters which are transmitted by an edge side integrated computing server and are trained by utilizing a U-net algorithm;
s2.2.3, receiving the new model parameters transmitted by the edge side integrated computing server, turning to step S2.2.2, replacing the original model parameters with the new model parameters, waiting for processing the digital image of the pathological section after next preprocessing, and outputting the result of the next preliminary diagnosis.
Step S6 is further included after step S1 to step S5, and specifically includes the following steps:
step S6: the method comprises the steps that a plurality of hospitals use the same edge side integrated computing server, the edge side integrated computing server receives digital pathological section digital images which are transmitted from each hospital system and are diagnosed by doctors, the images are divided into a training set, a verification set and a testing set, the training set and the testing set are input into an established network model, the model is trained again on the basis of the previous training weight, new model parameters are output, and the new model parameters are transmitted to edge side computing terminals of the hospitals.
The beneficial effect that this application reached:
the invention provides a pathological diagnosis system and method based on an edge side computing device, which can be used for conveniently storing, transmitting, computing and other operations on a digital image by combining an edge side computing technology and a digital scanner, identifying a focus and performing prescreening diagnosis while scanning and storing, and can be butted with a doctor diagnosis platform and a remote consultation system, so that the pathological section identification efficiency is greatly improved.
According to the invention, the edge side computing technology is adopted to quickly preprocess the digital pathological section, so as to assist doctors to accurately judge, and a joint learning mechanism is also adopted, so that the privacy of patients is protected, meanwhile, training results of multiple hospital workstations are collected, and corresponding correction parameters are returned. With the increase of pathological section data of different hospitals, the system can obtain a model with high robustness, high identification efficiency and high identification precision, and optimize and assist doctors to diagnose.
Drawings
Fig. 1 is a system configuration diagram of a pathology diagnosis system based on an edge side calculation and service apparatus according to an embodiment of the present invention, which is directed to a hospital;
FIG. 2 is a block diagram of a multi-hospital joint training system according to an embodiment of the present invention;
FIG. 3 is a diagram of an overall architecture of a multi-hospital joint training system according to an embodiment of the present invention.
Detailed Description
The present application is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present application is not limited thereby.
The invention provides a pathological diagnosis system and method based on an edge side calculation and service device. A pathological diagnosis system based on edge-side calculation and service device, when applied in a hospital, as shown in fig. 1, comprising: the system comprises a digital slice scanner, an edge side computing terminal, a doctor diagnosis workstation and an edge side integrated computing server;
the digital slice scanner is connected with the edge side computing terminal, and the edge side computing terminal is respectively connected with the doctor diagnosis workstation and the edge side integrated server;
the digital slice scanner is used for scanning pathological sections into pathological section digital images and transmitting the pathological section digital images to the edge side computing terminal;
the edge side computing terminal receives the pathological section digital image transmitted by the digital section scanner, carries out edge computing processing on the pathological section digital image, gives a preliminary diagnosis result comprising a focus part identification result and a segmentation result, and transmits the pathological section digital image and the preliminary diagnosis result after the edge computing processing to a doctor diagnosis workstation; meanwhile, receiving a doctor diagnosis result fed back by the doctor diagnosis workstation, confirming the marked part of the doctor diagnosis result, synthesizing the preliminary diagnosis result and the doctor diagnosis result into a test result, encrypting the test result and transmitting the encrypted test result to the edge side integrated computing server;
the doctor diagnosis workstation is used for carrying out medical diagnosis by a doctor according to the digital image of the pathological section after edge calculation and transmitting the diagnosis result of the doctor to the edge side calculation terminal;
the edge side integrated computing server: and receiving the encrypted test result transmitted by the edge side computing terminal, performing a new round of training on the basis of the previously trained model, and transmitting the model parameters after the new round of training to the edge side computing terminal.
The edge side computing terminal and the edge side integrated computing server are collectively called an edge side computing and service device.
The edge side integrated computing server is placed at a far end;
the edge side computing terminal is placed inside the hospital.
If the system is applied to multiple hospitals, as shown in fig. 2 and 3, the multiple hospitals adopt the same edge integrated computing server, the encrypted test results of all the hospitals are transmitted to the same edge integrated computing server, and the edge integrated computing server performs a new round of training again by using the encrypted test results of all the hospitals and transmits the model parameters after the new round of training to the edge computing terminals of all the hospitals.
And the edge side integrated computing server is transmitted to the edge side computing device in each hospital system through the secure link when a new round of training is performed.
The secure link, comprising:
(1) filtering out basic information of a patient;
(2) and carrying out encryption operation on the information in the secure link, wherein the secure link can be opened only through a secret key.
Firstly, carrying out hash operation on an original file; secondly, generating a random number to encrypt the original file; and finally, encrypting the random number by using the public key to ensure that only the edge side integrated computing server can open the secure link.
A pathological diagnosis method based on an edge side calculation and service device is realized by adopting a pathological diagnosis system based on the edge side calculation and service device, and comprises the following specific steps:
step S1: scanning the pathological section into a pathological section digital image by using a digital section scanner, and transmitting the pathological section digital image to the edge side computing terminal;
step S2: using an edge side computing terminal to receive the digital image of the pathological section transmitted by the digital section scanner, carrying out edge computing processing on the digital image of the pathological section, giving a preliminary diagnosis result comprising a lesion part identification result and a segmentation result, and transmitting the digital image of the pathological section and the preliminary diagnosis result after the edge computing processing to a doctor diagnosis workstation;
step S3: through the doctor diagnosis workstation, a doctor carries out medical diagnosis according to the digital image of the pathological section after edge calculation and transmits the diagnosis result of the doctor to an edge side calculation terminal;
step S4: integrating the doctor diagnosis result fed back by the doctor diagnosis workstation and the preliminary diagnosis result generated in the step S2 by using the edge side computing terminal to generate a test result, encrypting the test result and transmitting the encrypted test result to the edge side integrated computing server;
step S5: and carrying out a new round of model training by using the edge side integrated computing server according to the encrypted test result transmitted by the edge side computing terminal, and transmitting the model parameters after the new round of training to the edge side computing terminal.
In step S2, the specific steps are:
step S2.1: carrying out standardized preprocessing on the digital image of the pathological section;
step S2.2: and (4) performing edge calculation processing on the preprocessed digital image of the pathological section, and giving a preliminary diagnosis result.
The normalization pre-processing procedure is as follows:
step S2.1.1: normalizing the digital image of the pathological section;
in order to eliminate the interference of the light unevenness to the image, the image is normalized. That is, each pixel point of each channel of the image is divided by 255, so that each pixel point f (x, y) is limited to be between 0 and 1.
Step S2.1.2: sharpening the normalized digital image of the pathological section;
in order to make the image edge characteristics more obvious, the model identification is more accurate. The invention sharpens the image by using the Laplacian operator. I.e. convolving each 3 x 3 pixel of the image with a laplacian convolution kernel.
Step S2.1.3: and cutting the sharpened digital image into a specified size.
Since the pathological section image is too large, the embodiment of the present application cuts the sharpened image into a 1024 × 1024-sized image using a sliding window.
In the step 2.2, specifically:
s2.2.1: setting initial model parameters;
s2.2.2: placing the preprocessed pathological section digital images into a U-net network for testing, and outputting a preliminary diagnosis result, namely a plurality of images with pathological part identification images and segmented images; the U-net network is an initial network model and model parameters which are transmitted by an edge side integrated computing server and are trained by utilizing a U-net algorithm;
the U-net algorithm and the specific process of establishing the U-net network are both in the prior art, and are not described in detail in the application.
S2.2.3, receiving the new model parameters transmitted by the edge side integrated computing server, turning to step S2.2.2, replacing the original model parameters with the new model parameters, waiting for processing the digital image of the pathological section after next preprocessing, and outputting the result of the next preliminary diagnosis.
Step S6 is further included after step S1 to step S5, and specifically includes the following steps:
step S6: the method comprises the steps that a plurality of hospitals use the same edge side integrated computing server, the edge side integrated computing server receives digital pathological section digital images which are transmitted from each hospital system and diagnosed by doctors, the images are divided into a training set, a verification set and a testing set, the training set and the testing set are input into an established network model, the model is retrained on the basis of the previous training weight, evaluation indexes of the corresponding divided or classified model in F1-Score, ROC curve, PR curve and mAP detection indexes are selected according to a testing result to be evaluated, when the evaluation Score exceeds a set threshold value (generally determined by medical experts), the training is stopped, new model parameters are output, and the new model parameters are transmitted to edge side computing terminals of the hospitals.
The four detection indexes of F1-Score, ROC curve, PR curve and mAP are all in the prior art, and are not repeated in the application.
(1) The cut pictures transmitted from each hospital are divided into a training set of 60%, a validation set of 20% and a test set of 20%. (2) And inputting the training set and the test set into the network, and retraining the model on the basis of the previous training weight. And stopping training until the effect tested by the test set is better.
The present applicant has described and illustrated embodiments of the present invention in detail with reference to the accompanying drawings, but it should be understood by those skilled in the art that the above embodiments are merely preferred embodiments of the present invention, and the detailed description is only for the purpose of helping the reader to better understand the spirit of the present invention, and not for limiting the scope of the present invention, and on the contrary, any improvement or modification made based on the spirit of the present invention should fall within the scope of the present invention.
Claims (10)
1. A pathological diagnosis system based on an edge side calculation and service device is characterized in that,
the method comprises the following steps: the system comprises a digital slice scanner, an edge side computing terminal, a doctor diagnosis workstation and an edge side integrated computing server;
the digital slice scanner is connected with the edge side computing terminal, and the edge side computing terminal is respectively connected with the doctor diagnosis workstation and the edge side integrated server;
the digital slice scanner is used for scanning pathological sections into pathological section digital images and transmitting the pathological section digital images to the edge side computing terminal;
the edge side computing terminal receives the digital image of the pathological section transmitted by the digital section scanner, carries out edge computing processing on the digital image of the pathological section to give a preliminary diagnosis result, and transmits the digital image of the pathological section and the preliminary diagnosis result which are processed by the edge computing processing to a doctor diagnosis workstation; meanwhile, receiving a doctor diagnosis result fed back by the doctor diagnosis workstation, integrating the preliminary diagnosis result and the doctor diagnosis result into a test result, encrypting the test result and transmitting the encrypted test result to the edge side integrated computing server;
the doctor diagnosis workstation is used for carrying out medical diagnosis by a doctor according to the digital image of the pathological section after edge calculation and transmitting the diagnosis result of the doctor to the edge side calculation terminal;
the edge side integrated computing server: and receiving the encrypted test result transmitted by the edge side computing terminal, performing a new round of training on the basis of the previously trained model, and transmitting the model parameters after the new round of training to the edge side computing terminal.
2. An edge-side computing and services device-based pathology diagnosis system according to claim 1, wherein:
the edge side integrated computing server is placed at a far end;
the edge side computing terminal is placed inside the hospital.
3. An edge-side computing and services device-based pathology diagnosis system according to claim 1, wherein:
the edge side integrated computing server is characterized in that multiple hospitals adopt the same edge integrated computing server, encrypted test results of all the hospitals are transmitted to the same edge integrated computing server, the edge side integrated computing server conducts a new round of training again by using the encrypted test results of all the hospitals, and model parameters after the new round of training are transmitted to edge side computing terminals of all the hospitals.
4. An edge-side-computing-and-services-device-based pathology diagnosis system according to claim 3,
and the edge side integrated computing server is transmitted to the edge side computing device in each hospital system through the secure link when a new round of training is performed.
5. An edge-side-computing-and-services-device-based pathology diagnosis system according to claim 3,
the secure link, comprising:
(1) filtering out basic information of a patient;
(2) and carrying out encryption operation on the information in the secure link, wherein the secure link can be opened only through a secret key.
6. A pathological diagnosis method based on the edge side calculation and service device is realized by adopting the pathological diagnosis system based on the edge side calculation and service device of any one of claims 1 to 5, and comprises the following specific steps:
step S1: scanning the pathological section into a pathological section digital image by using a digital section scanner, and transmitting the pathological section digital image to the edge side computing terminal;
step S2: using an edge side computing terminal to receive the digital image of the pathological section transmitted by the digital section scanner, carrying out edge computing processing on the digital image of the pathological section, giving a preliminary diagnosis, and transmitting the digital image of the pathological section and a preliminary diagnosis result which are subjected to the edge computing processing to a doctor diagnosis workstation;
step S3: through the doctor diagnosis workstation, a doctor carries out medical diagnosis according to the digital image of the pathological section after edge calculation and transmits the diagnosis result of the doctor to an edge side calculation terminal;
step S4: integrating the doctor diagnosis result fed back by the doctor diagnosis workstation and the preliminary diagnosis result generated in the step S2 by using the edge side computing terminal to generate a test result, encrypting the test result and transmitting the encrypted test result to the edge side integrated computing server;
step S5: and carrying out a new round of model training by using the edge side integrated computing server according to the encrypted test result transmitted by the edge side computing terminal, and transmitting the model parameters after the new round of training to the edge side computing terminal.
7. An edge-side calculation and service apparatus-based pathology diagnosis method according to claim 6,
in step S2, the specific steps are:
step S2.1: carrying out standardized preprocessing on the digital image of the pathological section;
step S2.2: and (4) performing edge calculation processing on the preprocessed digital image of the pathological section, and giving a preliminary diagnosis result.
8. An edge-side calculation and service apparatus-based pathology diagnosis method according to claim 7,
the normalization pre-processing procedure is as follows:
step S2.1.1: normalizing the digital image of the pathological section;
step S2.1.2: sharpening the normalized digital image of the pathological section;
step S2.1.3: and cutting the sharpened digital image into a specified size.
9. An edge-side calculation and service apparatus-based pathology diagnosis method according to claim 6,
in the step 2.2, specifically:
s2.2.1: setting initial model parameters;
s2.2.2: placing the preprocessed pathological section digital images into a U-net network for testing, and outputting a preliminary diagnosis result, namely a plurality of images with pathological part identification and segmentation; the U-net network is an initial network model and model parameters which are transmitted by an edge side integrated computing server and are trained by utilizing a U-net algorithm;
s2.2.3, receiving the new model parameters transmitted by the edge side integrated computing server, turning to step S2.2.2, replacing the original model parameters with the new model parameters, waiting for processing the digital image of the pathological section after next preprocessing, and outputting the result of the next preliminary diagnosis.
10. An edge-side calculation and service apparatus-based pathology diagnosis method according to claim 6,
step S6 is further included after step S1 to step S5, and specifically includes the following steps:
step S6: the method comprises the steps that a plurality of hospitals use the same edge side integrated computing server, the edge side integrated computing server receives digital pathological section digital images which are transmitted from each hospital system and are diagnosed by doctors, the images are divided into a training set, a verification set and a testing set, the training set and the testing set are input into an established network model, the model is trained again on the basis of the previous training weight, new model parameters are output, and the new model parameters are transmitted to edge side computing terminals of the hospitals.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010344137.4A CN111370098A (en) | 2020-04-27 | 2020-04-27 | Pathological diagnosis system and method based on edge side calculation and service device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010344137.4A CN111370098A (en) | 2020-04-27 | 2020-04-27 | Pathological diagnosis system and method based on edge side calculation and service device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111370098A true CN111370098A (en) | 2020-07-03 |
Family
ID=71207456
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010344137.4A Pending CN111370098A (en) | 2020-04-27 | 2020-04-27 | Pathological diagnosis system and method based on edge side calculation and service device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111370098A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112162959A (en) * | 2020-10-15 | 2021-01-01 | 深圳技术大学 | Medical data sharing method and device |
WO2022077948A1 (en) * | 2020-10-14 | 2022-04-21 | 新智数字科技有限公司 | Federated learning-based internet of things data service method and apparatus, device, and medium |
CN117274185A (en) * | 2023-09-19 | 2023-12-22 | 阿里巴巴达摩院(杭州)科技有限公司 | Detection method, detection model product, electronic device, and computer storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040254503A1 (en) * | 2003-06-13 | 2004-12-16 | Sarvazyan Armen P. | Internet-based system and a method for automated analysis of tactile imaging data and detection of lesions |
WO2016192612A1 (en) * | 2015-06-02 | 2016-12-08 | 陈宽 | Method for analysing medical treatment data based on deep learning, and intelligent analyser thereof |
CN108288506A (en) * | 2018-01-23 | 2018-07-17 | 雨声智能科技(上海)有限公司 | A kind of cancer pathology aided diagnosis method based on artificial intelligence technology |
CN109598294A (en) * | 2018-11-23 | 2019-04-09 | 哈尔滨工程大学 | Cloud retina OCT identification intelligent diagnostic system and its application method based on hardware and software platform |
CN211654320U (en) * | 2020-04-27 | 2020-10-09 | 北京贝叶科技有限公司 | Pathological diagnosis system based on edge side calculation and service device |
-
2020
- 2020-04-27 CN CN202010344137.4A patent/CN111370098A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040254503A1 (en) * | 2003-06-13 | 2004-12-16 | Sarvazyan Armen P. | Internet-based system and a method for automated analysis of tactile imaging data and detection of lesions |
WO2016192612A1 (en) * | 2015-06-02 | 2016-12-08 | 陈宽 | Method for analysing medical treatment data based on deep learning, and intelligent analyser thereof |
CN108288506A (en) * | 2018-01-23 | 2018-07-17 | 雨声智能科技(上海)有限公司 | A kind of cancer pathology aided diagnosis method based on artificial intelligence technology |
CN109598294A (en) * | 2018-11-23 | 2019-04-09 | 哈尔滨工程大学 | Cloud retina OCT identification intelligent diagnostic system and its application method based on hardware and software platform |
CN211654320U (en) * | 2020-04-27 | 2020-10-09 | 北京贝叶科技有限公司 | Pathological diagnosis system based on edge side calculation and service device |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022077948A1 (en) * | 2020-10-14 | 2022-04-21 | 新智数字科技有限公司 | Federated learning-based internet of things data service method and apparatus, device, and medium |
CN112162959A (en) * | 2020-10-15 | 2021-01-01 | 深圳技术大学 | Medical data sharing method and device |
CN112162959B (en) * | 2020-10-15 | 2023-10-10 | 深圳技术大学 | Medical data sharing method and device |
CN117274185A (en) * | 2023-09-19 | 2023-12-22 | 阿里巴巴达摩院(杭州)科技有限公司 | Detection method, detection model product, electronic device, and computer storage medium |
CN117274185B (en) * | 2023-09-19 | 2024-05-07 | 阿里巴巴达摩院(杭州)科技有限公司 | Detection method, detection model product, electronic device, and computer storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111370098A (en) | Pathological diagnosis system and method based on edge side calculation and service device | |
CN109670510B (en) | Deep learning-based gastroscope biopsy pathological data screening system | |
CN111127412B (en) | Pathological image recognition device based on generation countermeasure network | |
KR102498686B1 (en) | Systems and methods for analyzing electronic images for quality control | |
Johnson et al. | Digital capture of fingerprints in a disaster victim identification setting: a review and case study | |
CN110516759B (en) | Soft tissue sarcoma transfer risk prediction system based on machine learning | |
CN211654320U (en) | Pathological diagnosis system based on edge side calculation and service device | |
CN113450305B (en) | Medical image processing method, system, equipment and readable storage medium | |
Ur Rehman et al. | Diabetic retinopathy fundus image classification using discrete wavelet transform | |
CN117711578A (en) | Medical image data analysis management system | |
US11967070B2 (en) | Systems and methods for automated image analysis | |
US20110075938A1 (en) | Identifying image abnormalities using an appearance model | |
CN109829921B (en) | Method and system for processing CT image of head, equipment and storage medium | |
Devi et al. | Automated dental identification system: An aid to forensic odontology | |
CN116543154A (en) | Medical image segmentation method based on multi-level semantic features | |
Smelyakov et al. | Lung X-Ray Images Preprocessing Algorithms for COVID-19 Diagnosing Intelligent Systems. | |
CN117716397A (en) | System and method for processing images for image matching | |
CN113796850A (en) | Parathyroid MIBI image analysis system, computer device, and storage medium | |
CN112185495A (en) | Medical equipment case data acquisition method and system | |
Kaddour et al. | Artificial Intelligence Pathologist: The use of Artificial Intelligence in Digital Healthcare | |
WO2020198560A1 (en) | Systems and methods for non-destructive de-identification of facial data in medical images | |
RU2697733C1 (en) | System for processing x-ray images and outputting result to user | |
CN117351196B (en) | Image segmentation method, device, computer equipment and storage medium | |
Khattar et al. | Investigation of Denoising Techniques for Removal of Hair and Noise from Dermoscopic Images | |
Sornil et al. | Predicting Diabetic Retinopathy Severity with Deep Learning: A Survey of Fundus Image Analysis Technique |
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 |