Intelligent image diagnosis system for radiology department

Abstract

The invention provides an intelligent image diagnosis system for the radiology department, and relates to the field of medical images. The intelligent image diagnosis system for the radiology department comprises a data acquisition module used for acquiring medical image data of a patient, including different types of images of X-ray, CT and MR I; the preprocessing module is used for preprocessing the collected image data so as to improve the image quality; the focus detection module is used for performing focus detection on the preprocessed image by using a deep learning algorithm, and automatically identifying and marking a focus area; and the diagnosis auxiliary module is used for performing intelligent diagnosis analysis and providing diagnosis suggestions and reports based on the focus detection result in combination with a medical image database and an expert knowledge base. By integrating a plurality of functional modules such as data acquisition, preprocessing, focus detection, intelligent diagnosis, user interaction, data storage and management and the like, a complete intelligent image diagnosis solution is formed, the problems of single function and scattered modules in the prior art are avoided, and the overall efficiency and reliability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging technology, and in particular to an intelligent imaging diagnosis system for radiology. Background Art

[0002] The radiology department is an important auxiliary examination department in the hospital. In the construction of modern hospitals, the radiology department is a department that integrates examination, diagnosis and treatment. Many diseases in various clinical departments must be examined by radiology equipment to achieve clear diagnosis and auxiliary diagnosis. The equipment of the radiology department generally includes ordinary X-ray machines, computer radiography systems (CR), direct digital radiography systems (DR), computer tomography (CT), magnetic resonance imaging (MRI), and digital subtraction angiography systems (DSA). Medical imaging technology (such as X-rays, CT, MRI, ultrasound, etc.) plays a vital role in modern medicine and is an important basis for disease diagnosis and treatment. Radiologists can discover lesions, evaluate the condition and formulate treatment plans by analyzing medical images. However, with the continuous updating of imaging equipment and the explosive growth of imaging data, radiologists are facing tremendous work pressure.

[0003] In recent years, the application of artificial intelligence technology in medical image analysis has gradually increased, providing new ideas and methods for solving the above problems. Artificial intelligence technology, especially deep learning algorithms, has shown strong capabilities in medical image processing. Although artificial intelligence technology has shown great potential in medical image analysis, there are still some problems with the existing artificial intelligence image diagnosis system. Most of the existing systems have single functions and lack integration between systems, making it difficult to achieve data sharing and process optimization. At the same time, the existing systems have limited diagnostic capabilities for complex cases and are prone to misdiagnosis or missed diagnosis. In addition, medical image data involves patient privacy, and the existing systems have security risks in data storage and management. Therefore, technical personnel in this field provide a radiology intelligent image diagnosis system to solve the problems raised in the above background technology. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the deficiencies of the prior art, the present invention provides an intelligent imaging diagnosis system for the radiology department, which solves the problems of the existing system having single functions, lack of integration between systems, and potential safety hazards in data storage and management.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A radiology intelligent imaging diagnosis system, comprising:

[0008] Data acquisition module, used to obtain medical imaging data of patients, including different types of images such as X-ray, CT and MRI;

[0009] The preprocessing module performs preprocessing on the collected image data, including denoising, normalization and image enhancement operations to improve the image quality;

[0010] The lesion detection module uses deep learning algorithms to detect lesions in preprocessed images, automatically identifying and marking lesion areas;

[0011] The diagnostic assistance module performs intelligent diagnostic analysis based on lesion detection results, combined with the medical imaging database and expert knowledge base, and provides diagnostic suggestions and reports;

[0012] The user interaction module provides a user-friendly interface for doctors to view images, diagnostic results and reports, and to conduct manual review and modification;

[0013] The data storage and management module stores and manages patients' imaging data, diagnostic results and reports, and supports data retrieval and statistical analysis.

[0014] Preferably, the data acquisition module includes a PACS interface module, a data receiving module, a data parsing module, a data storage module and a data management module, wherein:

[0015] The PACS interface module includes configuring PACS server information and establishing DICOM connection. The doctor enters the IP address, port number and AETitile information of the PACS server in the system configuration interface. The system stores the PACS server information in the configuration file. The system uses the configured PACS server information to establish a connection with the PACS server through the DICOM protocol. After the connection is successful, the system can send C-FIND and C-MOVE DICOM commands to the PACS server.

[0016] The data receiving module receives the image data transmitted from the PACS system. The doctor enters the basic information of the patient on the doctor's workstation. The system encapsulates the patient information into DICOM and C-FIND requests and sends them to the PACS server. The PACS server returns the query results, including all the image data information of the patient. The data receiving module receives the image data and passes it to the data parsing module.

[0017] Data parsing module: The data parsing module reads the received image data, parses the DICOM file, and extracts the image pixel data, image parameters, patient information, and image information;

[0018] Data storage module, the storage module stores the parsed image data in a local database. The database adopts a relational database or a NoSQL database. The image data is stored in DICOM format. Image parameters, patient information and image information are stored in the database table;

[0019] The data management module manages the stored image data, including data retrieval, data backup and data deletion.

[0020] Preferably, the preprocessing module includes a data reading module, a denoising processing module, a normalization processing module, an image enhancement module and a data output module, wherein:

[0021] The data reading module reads the stored image data from the database. The image data is stored in DICOM format. The data reading module supports reading in DICOM format.

[0022] The denoising processing module removes image noise by denoising the image data, and passes the denoised image data to the normalization processing module. The specific algorithm formula is as follows:

[0023]

[0024] Where G(x, y) is the Gaussian kernel function and σ is the standard deviation;

[0025] The normalization processing module normalizes the pixel values ​​to the range of 0-1 by normalizing the image data. The specific calculation formula is:

[0026]

[0027] Among them, I norm is the normalized pixel value, I is the original pixel value, and I min and I max are the minimum and maximum pixel values ​​respectively;

[0028] Image enhancement module, which improves image contrast, brightness and sharpness by performing image enhancement processing on image data;

[0029] The data output module outputs the preprocessed image data to the lesion detection module.

[0030] Preferably, the lesion detection module includes a data input module, a model loading module, a lesion detection reasoning module, a result output module and a result visualization module, wherein:

[0031] The data input module receives the preprocessed image data from the preprocessing module and performs necessary preprocessing operations on the received image data;

[0032] The model loading module loads the trained deep learning model from the model library. The deep learning model can adopt different network structures and load the model parameters, including convolution kernel weights and bias items.

[0033] The lesion detection inference module uses a deep learning model to detect lesions on image data. The model infers the input image data and outputs the lesion detection results. Based on the lesion detection results, the characteristics of the lesion, including the area, are calculated:

[0034]

[0035] Where N is the total number of pixels in the image, mask i is the value of the i-th pixel in the mask image;

[0036] Shape features:

[0037]

[0038] Among them, Circularity is the circularity, Compactness is the compactness, Eccentricity is the eccentricity, major axis and minor axis are the long axis and short axis lengths of the lesion respectively;

[0039] The result output module and the visualization module output the lesion detection results to the diagnosis auxiliary module through the result output module. The result visualization module visualizes the lesion detection results and marks the lesion area. The marking methods include: drawing a bounding box on the image, superimposing a mask image on the image, and displaying a heat map.

[0040] Preferably, the diagnosis auxiliary module includes a data input module, a feature extraction module, a knowledge base matching module, a diagnosis reasoning module, a report generation module and a result output module, wherein:

[0041] Data input module, receiving lesion detection results and basic patient information;

[0042] Feature extraction module, extracting lesion features and patient features;

[0043] The knowledge base matching module matches the extracted features with the medical imaging database and expert knowledge base;

[0044] The diagnostic reasoning module performs diagnostic reasoning and generates diagnostic suggestions based on the matching results;

[0045] The report generation module generates a diagnostic report and outputs diagnostic suggestions and reports through the result output module.

[0046] Preferably, the user interaction module includes a user interface module, an image display module, a diagnosis result display module, a report viewing module, a manual review module and a data management module, which provides a graphical user interface for doctors to operate, display the patient's medical imaging data, including original images and annotated images, display lesion detection results and diagnostic suggestions, and view the diagnostic report generated by the system. Doctors can manually review and modify the diagnostic results and manage the imaging data, diagnostic results and reports, including data retrieval and data export.

[0047] Preferably, the data storage and management module includes a data storage submodule, a data management submodule and a data security submodule, wherein:

[0048] Data storage submodule: responsible for storing imaging data, diagnostic results and reports in the database, supporting multiple database types;

[0049] Data management submodule: responsible for managing stored data, including data retrieval, data update, data deletion and data backup, and providing data access interface for other modules to call;

[0050] Data security submodule: responsible for ensuring data security, including data encryption, access control, data backup and recovery.

[0051] (III) Beneficial effects

[0052] The present invention provides an intelligent imaging diagnosis system for radiology, which has the following beneficial effects:

[0053] 1. The present invention integrates multiple functional modules such as data acquisition, preprocessing, lesion detection, intelligent diagnosis, user interaction, data storage and management, forming a complete intelligent imaging diagnosis solution, avoiding the problems of single function and scattered modules in the traditional method, and improving the overall efficiency and reliability of the system.

[0054] 2. In the present invention, the diagnostic assistance module performs intelligent diagnostic analysis based on the lesion detection results, combined with the medical imaging database and expert knowledge base, and provides diagnostic suggestions and reports. The system can simulate the doctor's diagnostic thinking process, help doctors make diagnostic decisions quickly, and improve diagnostic efficiency.

[0055] 3. In the present invention, the data storage and management module adopts advanced data security technology to ensure the security of patient data. The system supports functions such as data encryption, access control, data backup and recovery to ensure the security and reliability of data. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the overall system of the present invention;

[0057] Figure 2 It is a system flow diagram of the data acquisition module in the present invention;

[0058] Figure 3 It is a schematic diagram of the system flow of the pre-processing module in the present invention;

[0059] Figure 4 It is a schematic diagram of the system flow of the lesion detection module in the present invention;

[0060] Figure 5 It is a schematic diagram of the system flow of the diagnosis auxiliary module in the present invention;

[0061] Figure 6 It is a system flow diagram of the user interaction module in the present invention;

[0062] Figure 7 It is a system flow diagram of the data storage and management module in the present invention. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] Embodiment 1:

[0065] like Figure 1-7 As shown, an embodiment of the present invention provides an intelligent imaging diagnosis system for radiology, including:

[0066] Data acquisition module, used to obtain medical imaging data of patients, including different types of images such as X-ray, CT and MRI;

[0067] The preprocessing module performs preprocessing on the collected image data, including denoising, normalization and image enhancement operations to improve the image quality;

[0068] The lesion detection module uses deep learning algorithms to detect lesions in preprocessed images, automatically identifying and marking lesion areas;

[0069] The diagnostic assistance module performs intelligent diagnostic analysis based on lesion detection results, combined with the medical imaging database and expert knowledge base, and provides diagnostic suggestions and reports;

[0070] The user interaction module provides a user-friendly interface for doctors to view images, diagnostic results and reports, and to conduct manual review and modification;

[0071] The data storage and management module stores and manages patients' imaging data, diagnostic results and reports, and supports data retrieval and statistical analysis.

[0072] The data acquisition module includes a PACS interface module, a data receiving module, a data analysis module, a data storage module and a data management module, among which:

[0073] The PACS interface module includes configuring PACS server information and establishing DICOM connection. The doctor enters the IP address, port number and AETitile information of the PACS server in the system configuration interface. The system stores the PACS server information in the configuration file. The system uses the configured PACS server information to establish a connection with the PACS server through the DICOM protocol. After the connection is successful, the system can send C-FIND and C-MOVE DICOM commands to the PACS server.

[0074] The data receiving module receives the image data transmitted from the PACS system. The doctor enters the basic information of the patient on the doctor's workstation. The system encapsulates the patient information into DICOM and C-FIND requests and sends them to the PACS server. The PACS server returns the query results, including all the image data information of the patient. The data receiving module receives the image data and passes it to the data parsing module.

[0075] Data parsing module: The data parsing module reads the received image data, parses the DICOM file, and extracts the image pixel data, image parameters, patient information, and image information;

[0076] Data storage module, the storage module stores the parsed image data in a local database. The database adopts a relational database or a NoSQL database. The image data is stored in DICOM format. Image parameters, patient information and image information are stored in the database table;

[0077] The data management module manages the stored image data, including data retrieval, data backup and data deletion.

[0078] The preprocessing module includes a data reading module, a denoising processing module, a normalization processing module, an image enhancement module and a data output module, wherein:

[0079] The data reading module reads the stored image data from the database. The image data is stored in DICOM format. The data reading module supports reading in DICOM format.

[0080] The denoising processing module removes image noise by denoising the image data, and passes the denoised image data to the normalization processing module. The specific algorithm formula is as follows:

[0081]

[0082] Where G(x, y) is the Gaussian kernel function and σ is the standard deviation;

[0083] The normalization processing module normalizes the pixel values ​​to the range of 0-1 by normalizing the image data. The specific calculation formula is:

[0084]

[0085] Among them, I norm is the normalized pixel value, I is the original pixel value, and I min and I max are the minimum and maximum pixel values ​​respectively;

[0086] Image enhancement module, which improves image contrast, brightness and sharpness by performing image enhancement processing on image data;

[0087] The data output module outputs the preprocessed image data to the lesion detection module.

[0088] The lesion detection module includes a data input module, a model loading module, a lesion detection reasoning module, a result output module and a result visualization module, among which:

[0089] The data input module receives the preprocessed image data from the preprocessing module and performs necessary preprocessing operations on the received image data;

[0090] The model loading module loads the trained deep learning model from the model library. The deep learning model can adopt different network structures and load the model parameters, including convolution kernel weights and bias items.

[0091] The lesion detection inference module uses a deep learning model to detect lesions on image data. The model infers the input image data and outputs the lesion detection results. Based on the lesion detection results, the characteristics of the lesion, including the area, are calculated:

[0092]

[0093] Where N is the total number of pixels in the image, mask i is the value of the i-th pixel in the mask image;

[0094] Shape features:

[0095]

[0096] Among them, Circularity is the circularity, Compactness is the compactness, Eccentricity is the eccentricity, major axis and minor axis are the long axis and short axis lengths of the lesion respectively;

[0097] The result output module and the visualization module output the lesion detection results to the diagnosis auxiliary module through the result output module. The result visualization module visualizes the lesion detection results and marks the lesion area. The marking methods include: drawing a bounding box on the image, superimposing a mask image on the image, and displaying a heat map.

[0098] The diagnosis auxiliary module includes a data input module, a feature extraction module, a knowledge base matching module, a diagnosis reasoning module, a report generation module and a result output module, among which:

[0099] Data input module, receiving lesion detection results and basic patient information;

[0100] Feature extraction module, extracting lesion features and patient features;

[0101] The knowledge base matching module matches the extracted features with the medical imaging database and expert knowledge base;

[0102] The diagnostic reasoning module performs diagnostic reasoning and generates diagnostic suggestions based on the matching results;

[0103] The report generation module generates a diagnostic report and outputs diagnostic suggestions and reports through the result output module.

[0104] The user interaction module includes a user interface module, an image display module, a diagnosis result display module, a report viewing module, a manual review module and a data management module. It provides a graphical user interface for doctors to operate, display the patient's medical imaging data, including original images and annotated images, display lesion detection results and diagnostic suggestions, and view the diagnostic report generated by the system. Doctors manually review and modify the diagnostic results and manage the imaging data, diagnostic results and reports, including data retrieval and data export.

[0105] The data storage and management module includes a data storage submodule, a data management submodule and a data security submodule, wherein:

[0106] Data storage submodule: responsible for storing imaging data, diagnostic results and reports in the database, supporting multiple database types;

[0107] Data management submodule: responsible for managing stored data, including data retrieval, data update, data deletion and data backup, and providing data access interface for other modules to call;

[0108] Data security submodule: responsible for ensuring data security, including data encryption, access control, data backup and recovery.

[0109] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A radiology intelligent imaging diagnosis system, characterized by: include: Data acquisition module, used to obtain medical imaging data of patients, including different types of images such as X-ray, CT and MRI; The preprocessing module performs preprocessing on the collected image data, including denoising, normalization and image enhancement operations to improve the image quality; The lesion detection module uses deep learning algorithms to detect lesions in preprocessed images, automatically identifying and marking lesion areas; The diagnostic assistance module performs intelligent diagnostic analysis based on lesion detection results, combined with the medical imaging database and expert knowledge base, and provides diagnostic suggestions and reports; The user interaction module provides a user-friendly interface for doctors to view images, diagnostic results and reports, and to conduct manual review and modification; The data storage and management module stores and manages patients' imaging data, diagnostic results and reports, and supports data retrieval and statistical analysis.

2. The intelligent imaging diagnostic system for radiology according to claim 1, characterized in that: The data acquisition module includes a PACS interface module, a data receiving module, a data parsing module, a data storage module and a data management module, wherein: The PACS interface module includes configuring PACS server information and establishing DICOM connection. The doctor enters the IP address, port number and AETitile information of the PACS server in the system configuration interface. The system stores the PACS server information in the configuration file. The system uses the configured PACS server information to establish a connection with the PACS server through the DICOM protocol. After the connection is successful, the system can send C-FIND and C-MOVE DICOM commands to the PACS server. The data receiving module receives the image data transmitted from the PACS system. The doctor enters the basic information of the patient on the doctor's workstation. The system encapsulates the patient information into DICOM and C-FIND requests and sends them to the PACS server. The PACS server returns the query results, including all the image data information of the patient. The data receiving module receives the image data and passes it to the data parsing module. Data parsing module: The data parsing module reads the received image data, parses the DICOM file, and extracts the image pixel data, image parameters, patient information, and image information; Data storage module, the storage module stores the parsed image data in a local database. The database adopts a relational database or a NoSQL database. The image data is stored in DICOM format. Image parameters, patient information and image information are stored in the database table; The data management module manages the stored image data, including data retrieval, data backup and data deletion.

3. The intelligent imaging diagnostic system for radiology according to claim 1, characterized in that: The preprocessing module includes a data reading module, a denoising processing module, a normalization processing module, an image enhancement module and a data output module, wherein: The data reading module reads the stored image data from the database. The image data is stored in DICOM format. The data reading module supports reading in DICOM format. The denoising processing module removes image noise by denoising the image data, and passes the denoised image data to the normalization processing module. The specific algorithm formula is as follows: Where G(x, y) is the Gaussian kernel function and σ is the standard deviation; The normalization processing module normalizes the pixel values ​​to the range of 0-1 by normalizing the image data. The specific calculation formula is: Among them, I norm is the normalized pixel value, I is the original pixel value, and I min and I max are the minimum and maximum pixel values ​​respectively; Image enhancement module, which improves image contrast, brightness and sharpness by performing image enhancement processing on image data; The data output module outputs the preprocessed image data to the lesion detection module.

4. The intelligent imaging diagnostic system for radiology according to claim 1, characterized in that: The lesion detection module includes a data input module, a model loading module, a lesion detection reasoning module, a result output module and a result visualization module, wherein: The data input module receives the preprocessed image data from the preprocessing module and performs necessary preprocessing operations on the received image data; The model loading module loads the trained deep learning model from the model library. The deep learning model can adopt different network structures and load the model parameters, including convolution kernel weights and bias items. The lesion detection inference module uses a deep learning model to detect lesions on image data. The model infers the input image data and outputs the lesion detection results. Based on the lesion detection results, the characteristics of the lesion, including the area, are calculated: Where N is the total number of pixels in the image, mask i is the value of the i-th pixel in the mask image; Shape features: Among them, Circularity is the circularity, Compactness is the compactness, Eccentricity is the eccentricity, majoraxis and minor axis are the long axis and short axis lengths of the lesion respectively; The result output module and the visualization module output the lesion detection results to the diagnosis auxiliary module through the result output module. The result visualization module visualizes the lesion detection results and marks the lesion area. The marking methods include: drawing a bounding box on the image, superimposing a mask image on the image, and displaying a heat map.

5. The intelligent imaging diagnostic system for radiology according to claim 1, characterized in that: The diagnosis auxiliary module includes a data input module, a feature extraction module, a knowledge base matching module, a diagnosis reasoning module, a report generation module and a result output module, wherein: Data input module, receiving lesion detection results and basic patient information; Feature extraction module, extracting lesion features and patient features; The knowledge base matching module matches the extracted features with the medical imaging database and expert knowledge base; The diagnostic reasoning module performs diagnostic reasoning and generates diagnostic suggestions based on the matching results; The report generation module generates a diagnostic report and outputs diagnostic suggestions and reports through the result output module.

6. The intelligent imaging diagnostic system for radiology according to claim 1, characterized in that: The user interaction module includes a user interface module, an image display module, a diagnosis result display module, a report viewing module, a manual review module and a data management module. It provides a graphical user interface for doctors to operate, display the patient's medical imaging data, including original images and annotated images, display lesion detection results and diagnostic suggestions, and view the diagnostic report generated by the system. Doctors manually review and modify the diagnostic results and manage the imaging data, diagnostic results and reports, including data retrieval and data export.

7. The intelligent imaging diagnostic system for radiology according to claim 1, characterized in that: The data storage and management module includes a data storage submodule, a data management submodule and a data security submodule, wherein: Data storage submodule: responsible for storing imaging data, diagnostic results and reports in the database, supporting multiple database types; Data management submodule: responsible for managing stored data, including data retrieval, data update, data deletion and data backup, and providing data access interface for other modules to call; Data security submodule: responsible for ensuring data security, including data encryption, access control, data backup and recovery.

Images