CN113576508A - Cerebral hemorrhage auxiliary diagnosis system based on neural network - Google Patents
Cerebral hemorrhage auxiliary diagnosis system based on neural network Download PDFInfo
- Publication number
- CN113576508A CN113576508A CN202110826032.7A CN202110826032A CN113576508A CN 113576508 A CN113576508 A CN 113576508A CN 202110826032 A CN202110826032 A CN 202110826032A CN 113576508 A CN113576508 A CN 113576508A
- Authority
- CN
- China
- Prior art keywords
- image
- neural network
- cerebral hemorrhage
- module
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 206010008111 Cerebral haemorrhage Diseases 0.000 title claims abstract description 51
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 47
- 238000003745 diagnosis Methods 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 31
- 230000011218 segmentation Effects 0.000 claims abstract description 15
- 208000032843 Hemorrhage Diseases 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000004364 calculation method Methods 0.000 claims abstract description 7
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 230000008030 elimination Effects 0.000 claims abstract description 6
- 238000003379 elimination reaction Methods 0.000 claims abstract description 6
- 230000000877 morphologic effect Effects 0.000 claims abstract description 6
- 230000006835 compression Effects 0.000 claims description 5
- 238000007906 compression Methods 0.000 claims description 5
- 238000004088 simulation Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 3
- 208000008574 Intracranial Hemorrhages Diseases 0.000 claims 1
- 238000011282 treatment Methods 0.000 abstract description 8
- 238000000034 method Methods 0.000 abstract description 6
- 238000002591 computed tomography Methods 0.000 description 39
- 230000000740 bleeding effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000001154 acute effect Effects 0.000 description 4
- 201000010099 disease Diseases 0.000 description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 210000004556 brain Anatomy 0.000 description 2
- 208000026106 cerebrovascular disease Diseases 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 208000008116 Traumatic Cerebral Hemorrhage Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 210000005013 brain tissue Anatomy 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011369 optimal treatment Methods 0.000 description 1
- 238000010827 pathological analysis Methods 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000002269 spontaneous effect Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
- A61B6/032—Transmission computed tomography [CT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
- A61B6/501—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of the head, e.g. neuroimaging or craniography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5205—Devices using data or image processing specially adapted for radiation diagnosis involving processing of raw data to produce diagnostic data
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- 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
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Radiology & Medical Imaging (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Pathology (AREA)
- General Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Heart & Thoracic Surgery (AREA)
- Veterinary Medicine (AREA)
- High Energy & Nuclear Physics (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Neurosurgery (AREA)
- Neurology (AREA)
- Pulmonology (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses a cerebral hemorrhage auxiliary diagnosis system based on a neural network, which belongs to the technical field of medical image processing and segmentation and comprises the following components: the acquisition module is used for acquiring a plurality of original CT images and marking a focus area; the processing module is used for carrying out image enhancement on the original CT images to obtain a training set; the training module is used for training the neural network by utilizing the training set; the first diagnosis module is used for acquiring a CT image shot for the second time, preprocessing the CT image, inputting the processed image into a trained neural network, and obtaining a gray level map which is consistent with the size of the input image and represents the confidence coefficient of each pixel; the second diagnosis module is used for sequentially carrying out threshold segmentation, morphological closed operation and hole elimination processing on the gray-scale image; and realizing the detection of cerebral hemorrhage focus and the calculation of hemorrhage amount according to the processed gray level image. Therefore, the cerebral hemorrhage diagnosis and treatment method can improve the efficiency and accuracy of cerebral hemorrhage diagnosis and treatment.
Description
Technical Field
The invention belongs to the technical field of medical image processing and segmentation, and particularly relates to a cerebral hemorrhage auxiliary diagnosis system based on a neural network.
Background
The cerebral hemorrhage refers to primary non-traumatic cerebral parenchymal hemorrhage, also called spontaneous cerebral hemorrhage, is the disease type with the highest fatality rate in acute cerebrovascular diseases, and accounts for 20% -30% of the acute cerebrovascular diseases. At present, acute cerebrovascular-related diseases have become the third most fatal disease. The determination of the amount of cerebral hemorrhage can be used as an important reference for the treatment means adopted by the subsequent doctors.
At present, X-ray Computed Tomography (CT) is generally adopted clinically as a main screening means for cerebral hemorrhage, and has the advantages of convenience and rapidness in operation, economy, no wound and the like. However, because the bleeding area in the cerebral hemorrhage CT image is easy to be confused with other normal brain tissues, the difficulty is higher when a diagnostician qualitatively observes and quantitatively analyzes; meanwhile, the cerebral hemorrhage is acute and rapid in development, and the symptoms reach a peak within minutes or hours after the development. However, since the pathological diagnosis needs to be performed under a microscope, the workload of visual observation is large, the CT report generally needs 6 to 24 hours to be obtained, the waiting time is too long, and the examination report is easily influenced by the experience of a pathologist, the fatigue state and the like. Meanwhile, part of remote areas are limited by medical resource distribution, and especially doctors with inexperience are difficult to quickly and definitely diagnose by CT, so that patients easily miss the optimal treatment opportunity.
With the development of machine learning technology and the wide application thereof in the medical field, in the field of cerebral hemorrhage auxiliary diagnosis, the cerebral hemorrhage CT and pathological section image auxiliary analysis tool is established based on the machine learning method, and the strong image processing and matrix operation capability of a computer can be utilized, so that the image analysis efficiency is improved, the workload of doctors is reduced, and the regional medical resource imbalance can be relieved to a certain extent. However, in consideration of protecting privacy of patients, the CT images captured by the CT apparatus are stored in the internal system of the hospital, and generally only support printing of films, and it is difficult to directly open a data interface to an external program, so that the above-mentioned machine learning-based auxiliary analysis program is difficult to be applied to the original captured images. One way to solve this problem is to take a second shot of the printed CT film and use it as the input of the program, but the quality of the second shot CT image is greatly affected by the shooting angle of view and the lighting conditions, so that it is difficult for the conventional auxiliary analysis program to obtain an accurate diagnosis result of cerebral hemorrhage. Therefore, designing an effective cerebral hemorrhage auxiliary diagnosis system aiming at the secondary shooting CT image is a problem to be solved at present and has important practical application value.
Disclosure of Invention
Aiming at the defects or the improvement requirements of the prior art, the invention provides the cerebral hemorrhage auxiliary diagnosis system based on the neural network, and the trained neural network can be used for completing the detection of the hemorrhage focus, the calculation of the amount of hemorrhage and the like by only shooting the CT image of the patient by using a mobile phone for uploading, so that a doctor is helped to determine the cerebral hemorrhage condition of the patient, and the diagnosis efficiency is improved.
In order to achieve the above object, the present invention provides a cerebral hemorrhage auxiliary diagnosis system based on a neural network, comprising the following modules:
the acquisition module is used for acquiring a plurality of original CT images and marking a focus area;
the processing module is used for carrying out image enhancement on the original CT images to obtain a training set;
the training module is used for training the neural network by utilizing the training set;
the first diagnosis module is used for acquiring a CT image shot for the second time, preprocessing the CT image, inputting the processed image into a trained neural network, and obtaining a gray level map which is consistent with the size of the input image and represents the confidence coefficient of each pixel;
the second diagnosis module is used for sequentially carrying out threshold segmentation, morphological closed operation and hole elimination processing on the gray-scale image; and realizing the detection of cerebral hemorrhage focus and the calculation of hemorrhage amount according to the processed gray level image.
Further, the training module is further configured to introduce the multiple adjacent images into an input layer of the neural network for joint training.
Further, the training module is further configured to calculate a loss between the gray-scale image output by the neural network and the labeled original CT image according to the selected cross entropy loss function, and apply back propagation of the loss to update the neural network weight parameter.
Further, acquiring and preprocessing a CT image of the secondary photographing, including: and acquiring a CT image of secondary shooting, and performing image center cutting and image compression to enable the processed image to be consistent with the original CT image in shape and size.
Further, after the image center cropping and the image compression are performed, the first diagnosis module is further used for performing graying processing on the image.
Further, the means of image enhancement employed includes: random dithering of gray scale, contrast adjustment, simulation of various noises, filtering operation and superposition of random bright spots.
Further, the neural network is a full convolution semantic segmentation neural network.
Generally, by the above technical solution conceived by the present invention, the following beneficial effects can be obtained:
(1) in the cerebral hemorrhage auxiliary diagnosis system provided by the invention, the original CT image is selected to be secondarily shot by user equipment such as a mobile phone and the like in consideration that in practical application, the CT image shot by the CT device cannot be immediately exported, and the bleeding focus detection, the bleeding amount calculation and the like cannot be quickly completed by utilizing the trained neural network. And because the image of the secondary shooting is different from the original CT image, the characteristics of the gray distribution, the image shape and the like of the original CT image in the training set are changed by an image enhancement method to be closer to the characteristics of the image of the secondary shooting, thereby improving the generalization capability of the learned model. After obtaining the corresponding gray-scale map based on the trained neural network, the method carries out threshold segmentation, morphological closing operation and hole elimination treatment in sequence, and restores the real cerebral hemorrhage condition. Therefore, the cerebral hemorrhage diagnosis and treatment method can improve the efficiency and accuracy of cerebral hemorrhage diagnosis and treatment.
(2) Before the images of the training set are input into the neural network, a program for importing a plurality of adjacent images together is designed, the images are firstly sorted according to the sequence of the images, before the current image is imported into the neural network, the adjacent images before and after the current image are inquired, and the adjacent images and the current image are imported into an input layer of the neural network together for joint training. Therefore, the false alarm rate of the final result can be reduced, and the credibility of the final segmentation result is higher.
Drawings
Fig. 1 is a block diagram of a neural network-based cerebral hemorrhage auxiliary diagnosis system according to an embodiment of the present invention;
fig. 2 is a diagram illustrating visualization effect of cerebral hemorrhage detection provided by the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
In the present application, the terms "first," "second," and the like (if any) in the description and the drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
As shown in fig. 1, the neural network-based cerebral hemorrhage auxiliary diagnosis system 100 provided by the present invention includes:
an obtaining module 110, configured to obtain multiple original CT images and mark a lesion area;
specifically, JSON label files marked by professionals are converted into masks input by the model in batches, namely jpg binary images with the sizes consistent with those of CT images. And inputting Mask as correctly labeled data into the model, wherein the two values of the Mask respectively represent the focus area with cerebral hemorrhage and the focus area without cerebral hemorrhage.
A processing module 120, configured to perform image enhancement on the multiple original CT images to obtain a training set;
specifically, the gray-scale image of the original CT is directly obtained, but the CT image obtained by secondary shooting by the user equipment such as a mobile phone is actually used by the system, so that the distribution of the training set data and the test set data is inconsistent for the neural network. For the problem, after analyzing image histograms of an original CT image and a secondary-shot CT image, a reasonable image enhancement method is selected in a targeted manner and is used for changing characteristics such as gray level distribution, image shape and the like of the original CT image in a training set to enable the characteristics to be closer to those of the secondary-shot image, and therefore the generalization ability of the learned model is improved. The means of image enhancement employed include: random dithering of gray scale, contrast adjustment (selecting proper contrast parameters for a histogram of a secondary shot image), simulation of various noises (noise interference in simulation shooting), filtering operation (dithering blur possibly generated in simulation shooting), superposition of random bright spots (used for simulating reflection in shooting).
A training module 130 for training a neural network using the training set;
specifically, the neural network is a full convolution semantic segmentation neural network, and the network has a remarkable effect on the segmentation problem of the medical image; the network mainly comprises an encoder (encoder), a decoder (decoder) and a skip connection (skip connection): the coder is used for down-sampling and converting the image into a feature map (feature map) containing high semantic information; the decoder acts as an upsampling, converting high semantic information into a score map (score map) of pixel level classification; and the skip layer connection is used for connecting the front characteristic diagram and the rear characteristic diagram under the same scale. The output of the network is a grayscale map representing the confidence of each pixel, consistent with the input image size. And calculating the loss of the gray-scale image output by the neural network and the marked original CT image according to the selected cross entropy loss function, and reversely propagating the loss for updating the weight parameters of the neural network.
Further, before the images of the training set are input into the neural network, a program for collectively importing a plurality of adjacent images is designed, the images are firstly sorted according to the sequence of the images, before the current image is imported into the neural network, the front and rear adjacent images of the current image are inquired, and are imported into an input layer of the neural network together with the current image to perform joint training. Therefore, the false alarm rate of the final result can be reduced, and the credibility of the final segmentation result is higher.
The first diagnosis module 140 is configured to acquire a CT image obtained by secondary photographing, perform preprocessing, input the processed image into a trained neural network, and obtain a gray scale map representing a confidence of each pixel, which is consistent with the size of the input image;
specifically, a doctor stably shoots and uploads a CT image with high precision under a good illumination condition; acquiring a secondary-shot CT image, and preprocessing the image, wherein the preprocessing comprises the following steps: image center cropping (cropping to a square shape consistent with the training set image), image compression (compressing the image to a size consistent with the training set, reducing the amount of computation in the testing process), and image graying.
The second diagnosis module 150 is used for sequentially performing threshold segmentation, morphological closed operation and hole elimination processing on the gray-scale image; and realizing the detection of cerebral hemorrhage focus and the calculation of hemorrhage amount according to the processed gray level image.
Specifically, threshold segmentation processing (primarily obtaining a foreground representing a cerebral hemorrhage focus and a background representing no focus), morphological closing operation (used for eliminating isolated points after segmentation and fusing narrow connected regions), and hole elimination operation (eliminating holes in the focus foreground by searching the connected regions and restoring the real cerebral hemorrhage condition) are performed on the gray level image to obtain a finally predicted cerebral hemorrhage image Mask. Calculating the blood volume according to a patient cerebral hemorrhage image Mask, judging whether cerebral hemorrhage occurs, judging cerebral hemorrhage when the cerebral hemorrhage volume is larger than a set error acceptable range, and outputting the cerebral hemorrhage volume.
In addition, a packaged auxiliary diagnosis small program can be erected at a WeChat end of user equipment such as a mobile phone, and after the small program is started, the WeChat end of the mobile phone outputs the calculation results of the yes/no cerebral hemorrhage and the amount of bleeding of the patient according to the interactively input basic information of the patient and the CT image content of the secondary shooting and the trained model; according to the interactive region selection of the small program page, the information of different bleeding point positions, bleeding amount and the like of the brain CT can be amplified manually, and then a preliminary diagnosis conclusion is output.
The visual effect diagram for detecting the cerebral hemorrhage provided by the embodiment of the invention is shown in fig. 2, can accurately position a hemorrhage focus, and in an identified image, a white area is an area with the cerebral hemorrhage focus, a black area is an area without the cerebral hemorrhage focus, and a conclusion of yes/no cerebral hemorrhage and a hemorrhage amount is output at the same time.
In practical application, the invention can carry out linkage diagnosis on a plurality of diseases of the brain simultaneously to form an all-round comprehensive diagnosis and treatment result, breaks through the limitation that each department of the existing medicine carries out independent diagnosis and treatment and has a limited range, reduces the probability of repeated inquiry and examination, and has more accurate and comprehensive diagnosis.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (7)
1. A brain hemorrhage auxiliary diagnosis system based on a neural network is characterized by comprising the following modules:
the acquisition module is used for acquiring a plurality of original CT images and marking a focus area;
the processing module is used for carrying out image enhancement on the original CT images to obtain a training set;
the training module is used for training the neural network by utilizing the training set;
the first diagnosis module is used for acquiring a CT image shot for the second time, preprocessing the CT image, inputting the processed image into a trained neural network, and obtaining a gray level map which is consistent with the size of the input image and represents the confidence coefficient of each pixel;
the second diagnosis module is used for sequentially carrying out threshold segmentation, morphological closed operation and hole elimination processing on the gray-scale image; and realizing the detection of cerebral hemorrhage focus and the calculation of hemorrhage amount according to the processed gray level image.
2. The neural-network-based cerebral hemorrhage aided diagnosis system as claimed in claim 1, wherein the training module is further configured to introduce multiple adjacent images into an input layer of the neural network for joint training.
3. The neural network-based cerebral hemorrhage aided diagnosis system as claimed in claim 1 or 2, wherein the training module is further configured to calculate a loss of the gray scale map of the neural network output and the labeled original CT image according to the selected cross entropy loss function, and to use the back propagation of the loss for updating the neural network weight parameters.
4. The neural network-based cerebral hemorrhage auxiliary diagnostic system according to claim 1, wherein the acquiring and preprocessing of the secondary captured CT image comprises: and acquiring a CT image of secondary shooting, and performing image center cutting and image compression to enable the processed image to be consistent with the original CT image in shape and size.
5. The neural network-based cerebral hemorrhage auxiliary diagnostic system according to claim 4, wherein the first diagnostic module is further configured to perform graying processing on the image after performing image center cropping and image compression.
6. The neural network-based cerebral hemorrhage aided diagnosis system according to claim 1, wherein the image enhancement means comprises: random dithering of gray scale, contrast adjustment, simulation of various noises, filtering operation and superposition of random bright spots.
7. The neural network-based cerebral hemorrhage aided diagnosis system according to claim 1, wherein the neural network is a full convolution semantic segmentation neural network.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110826032.7A CN113576508B (en) | 2021-07-21 | 2021-07-21 | Cerebral hemorrhage auxiliary diagnosis system based on neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110826032.7A CN113576508B (en) | 2021-07-21 | 2021-07-21 | Cerebral hemorrhage auxiliary diagnosis system based on neural network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113576508A true CN113576508A (en) | 2021-11-02 |
CN113576508B CN113576508B (en) | 2024-07-09 |
Family
ID=78248798
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110826032.7A Active CN113576508B (en) | 2021-07-21 | 2021-07-21 | Cerebral hemorrhage auxiliary diagnosis system based on neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113576508B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114049938A (en) * | 2021-11-22 | 2022-02-15 | 上海商汤智能科技有限公司 | Image detection method and related device, electronic equipment and storage medium |
CN114549532A (en) * | 2022-04-27 | 2022-05-27 | 珠海市人民医院 | Cerebral ischemia auxiliary analysis method and system based on medical image processing |
CN115274099A (en) * | 2022-09-26 | 2022-11-01 | 之江实验室 | Human-intelligent interactive computer-aided diagnosis system and method |
CN116309647A (en) * | 2023-04-27 | 2023-06-23 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109949309A (en) * | 2019-03-18 | 2019-06-28 | 安徽紫薇帝星数字科技有限公司 | A kind of CT image for liver dividing method based on deep learning |
CN110503630A (en) * | 2019-07-19 | 2019-11-26 | 江苏师范大学 | A kind of cerebral hemorrhage classification, positioning and prediction technique based on three dimensional depth learning model |
CN111242925A (en) * | 2020-01-13 | 2020-06-05 | 北京妙医佳健康科技集团有限公司 | Target detection method and device for CT image data and electronic equipment |
CN111445457A (en) * | 2020-03-26 | 2020-07-24 | 北京推想科技有限公司 | Network model training method and device, network model identification method and device, and electronic equipment |
CN111833311A (en) * | 2020-06-18 | 2020-10-27 | 安徽农业大学 | Image identification method based on deep learning and application of image identification method to rice disease identification |
CN112686336A (en) * | 2021-01-28 | 2021-04-20 | 杭州电子科技大学 | Burn surface of a wound degree of depth classification system based on neural network |
CN112991320A (en) * | 2021-04-07 | 2021-06-18 | 德州市人民医院 | System and method for predicting hematoma expansion risk of cerebral hemorrhage patient |
CN113095470A (en) * | 2020-01-08 | 2021-07-09 | 字节跳动有限公司 | Neural network training method, image processing method and device, and storage medium |
-
2021
- 2021-07-21 CN CN202110826032.7A patent/CN113576508B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109949309A (en) * | 2019-03-18 | 2019-06-28 | 安徽紫薇帝星数字科技有限公司 | A kind of CT image for liver dividing method based on deep learning |
CN110503630A (en) * | 2019-07-19 | 2019-11-26 | 江苏师范大学 | A kind of cerebral hemorrhage classification, positioning and prediction technique based on three dimensional depth learning model |
CN113095470A (en) * | 2020-01-08 | 2021-07-09 | 字节跳动有限公司 | Neural network training method, image processing method and device, and storage medium |
CN111242925A (en) * | 2020-01-13 | 2020-06-05 | 北京妙医佳健康科技集团有限公司 | Target detection method and device for CT image data and electronic equipment |
CN111445457A (en) * | 2020-03-26 | 2020-07-24 | 北京推想科技有限公司 | Network model training method and device, network model identification method and device, and electronic equipment |
CN111833311A (en) * | 2020-06-18 | 2020-10-27 | 安徽农业大学 | Image identification method based on deep learning and application of image identification method to rice disease identification |
CN112686336A (en) * | 2021-01-28 | 2021-04-20 | 杭州电子科技大学 | Burn surface of a wound degree of depth classification system based on neural network |
CN112991320A (en) * | 2021-04-07 | 2021-06-18 | 德州市人民医院 | System and method for predicting hematoma expansion risk of cerebral hemorrhage patient |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114049938A (en) * | 2021-11-22 | 2022-02-15 | 上海商汤智能科技有限公司 | Image detection method and related device, electronic equipment and storage medium |
WO2023087641A1 (en) * | 2021-11-22 | 2023-05-25 | 上海商汤智能科技有限公司 | Image detection method and related apparatus, electronic device, and storage medium |
CN114549532A (en) * | 2022-04-27 | 2022-05-27 | 珠海市人民医院 | Cerebral ischemia auxiliary analysis method and system based on medical image processing |
CN115274099A (en) * | 2022-09-26 | 2022-11-01 | 之江实验室 | Human-intelligent interactive computer-aided diagnosis system and method |
CN115274099B (en) * | 2022-09-26 | 2022-12-30 | 之江实验室 | Human-intelligent interactive computer-aided diagnosis system and method |
CN116309647A (en) * | 2023-04-27 | 2023-06-23 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device |
CN116309647B (en) * | 2023-04-27 | 2023-07-25 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device |
Also Published As
Publication number | Publication date |
---|---|
CN113576508B (en) | 2024-07-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110458883B (en) | Medical image processing system, method, device and equipment | |
CN110060774B (en) | Thyroid nodule identification method based on generative confrontation network | |
CN113576508A (en) | Cerebral hemorrhage auxiliary diagnosis system based on neural network | |
CN108464840B (en) | Automatic detection method and system for breast lumps | |
Dias et al. | Retinal image quality assessment using generic image quality indicators | |
CN112070119B (en) | Ultrasonic section image quality control method, device and computer equipment | |
CN111862044B (en) | Ultrasonic image processing method, ultrasonic image processing device, computer equipment and storage medium | |
TWI719587B (en) | Pre-processing method and storage device for quantitative analysis of fundus image | |
JPH07504056A (en) | Methods for identifying objects using data processing techniques | |
US11967181B2 (en) | Method and device for retinal image recognition, electronic equipment, and storage medium | |
CN113012163A (en) | Retina blood vessel segmentation method, equipment and storage medium based on multi-scale attention network | |
CN114450716A (en) | Image processing for stroke characterization | |
Ahmed et al. | Deep learning based automated detection of intraretinal cystoid fluid | |
CN114782337A (en) | OCT image recommendation method, device, equipment and medium based on artificial intelligence | |
CN114372985A (en) | Diabetic retinopathy focus segmentation method and system adapting to multi-center image | |
Lee et al. | Improved classification of brain-tumor mri images through data augmentation and filter application | |
CN111339993A (en) | X-ray image metal detection method and system | |
CN114612484B (en) | Retina OCT image segmentation method based on unsupervised learning | |
CN113537407B (en) | Image data evaluation processing method and device based on machine learning | |
Li et al. | A deep-learning-enabled monitoring system for ocular redness assessment | |
CN111462067B (en) | Image segmentation method and device | |
Rao et al. | OTONet: Deep Neural Network for Precise Otoscopy Image Classification | |
CN113796850A (en) | Parathyroid MIBI image analysis system, computer device, and storage medium | |
CN112734701A (en) | Fundus focus detection method, fundus focus detection device and terminal equipment | |
CN112967295A (en) | Image processing method and system based on residual error network and attention mechanism |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |