Medical image information intelligent management system

Abstract

The invention relates to the technical field of medical information, and discloses a medical image information intelligent management system, which comprises a data acquisition layer, a data storage layer, a data processing layer and an application layer, and is characterized in that the data acquisition layer comprises an image acquisition unit which is used for being connected with various medical image equipment through a standardized interface protocol and acquiring medical image data in real time; according to the intelligent management system for medical image information, an image enhancement sub-module in an image preprocessing module improves the overall contrast and detail definition of an image through histogram equalization and contrast limited adaptive histogram equalization technologies, so that a doctor can observe lesion features in the image more clearly, and the medical image information can be obtained more accurately. The noise reduction sub-module removes image noise, improves image quality and provides a more reliable image basis for subsequent diagnosis, and the image segmentation sub-module can accurately segment an abnormal region in a medical image by means of a deep learning algorithm and provides accurate image data for medical diagnosis.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and specifically to an intelligent management system for medical image information. Background Art

[0002] In today's medical field, medical images play an irreplaceable and crucial role in the accurate diagnosis of diseases, the formulation of reasonable treatment plans, and the effective tracking of the condition. From the early simple X-ray imaging to the current advanced imaging technologies such as CT, MRI, and PET, the types and quantities of medical images have shown an explosive growth. With the rapid development of emerging technologies such as artificial intelligence, big data, and cloud computing, the medical industry has put forward higher requirements for the management of medical images. With the emergence of new technologies, the intelligent management of medical images can be realized to improve the management efficiency and quality. On the other hand, it is expected to provide more valuable decision-making support for clinical diagnosis and treatment through the in-depth mining and analysis of massive medical image data. However, there are some defects in the existing storage of medical images, such as: The traditional medical image management method adopts a decentralized storage mode. The decentralized storage mode not only makes the unified management and maintenance of data extremely difficult, but also greatly increases the risk of data loss or damage. Moreover, each department uses different devices and different media for storage, which increases the difficulty of information sharing of image data. And due to the decentralized management of images, the difficulty of image retrieval increases, and medical staff need to spend a lot of time finding specific images.

[0003] In view of the above problems, there is an urgent need to innovate and design on the basis of the original medical image information management system. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent management system for medical image information to solve the problems raised in the above background art that the decentralized storage mode not only makes the unified management and maintenance of data extremely difficult, but also greatly increases the risk of data loss or damage, and each department uses different devices and different media for storage, which increases the difficulty of information sharing of image data.

[0005] To achieve the above purpose, the present invention provides the following technical solution: An intelligent management system for medical image information, including a data acquisition layer, a data storage layer, a data processing layer, and an application layer, where: The data acquisition layer includes an image acquisition unit, which is used to connect with various medical imaging devices through a standardized interface protocol, collect medical image data in real time, and has a data block transmission and caching function to ensure the stability and efficiency of data acquisition; The data storage layer has a distributed storage architecture, uses AES data encryption technology to encrypt and store medical image data, and is equipped with a key management system to ensure key security. It also includes an image retrieval and query module for quickly retrieving and querying stored medical images; The data processing layer is provided with an image preprocessing module for preprocessing the collected medical images, and an intelligent diagnosis module for using a deep learning model to perform intelligent diagnosis on the preprocessed images; The application layer provides an operation interface for medical staff, realizes functions such as querying, browsing, and comparative analysis of medical images, supports image browsing and interaction, displays the diagnostic results output by the data processing layer, and can interface with the hospital information system and the electronic medical record system to achieve information interconnection.

[0006] Adopting the above technical solution, through a centralized storage device using distributed storage, it is convenient to further centrally store different medical imaging data.

[0007] Preferably, the standardized interface protocol in the data acquisition layer is the DICOM protocol, which is compatible with various medical imaging devices and standardizes data transmission. The data items in the data acquisition layer are transported to the data processing layer.

[0008] Adopting the above technical solution is convenient for providing normalized data processing for the collected medical image data and convenient for medical staff to query images.

[0009] Preferably, the distributed storage architecture is based on the Ceph distributed storage system and uses erasure coding technology to achieve redundant data storage.

[0010] Adopting the above technical solution, using the storage method of a distributed storage rack is convenient for providing a classified storage effect for different medical images.

[0011] Preferably, the image preprocessing module includes an image enhancement sub-module, a noise reduction sub-module, and an image segmentation sub-module. The image enhancement sub-module uses histogram equalization and contrast-limited adaptive histogram equalization technologies to enhance the image contrast.

[0012] Adopting the above technical solution, the image enhancement sub-module, the noise reduction sub-module, and the image segmentation sub-module in the image preprocessing module provide corresponding image processing.

[0013] Preferably, the noise reduction sub-module uses algorithms such as Gaussian filtering and median filtering to remove image noise, and the image segmentation sub-module uses a deep learning algorithm to segment abnormal regions in medical images.

[0014] Adopting the above technical solution, the noise reduction sub-module provides corresponding processing for medical images.

[0015] Preferably, the deep learning model is a medical image diagnosis model constructed based on a convolutional neural network. The model parameters are initialized using transfer learning technology, and the model is trained using optimization algorithms such as stochastic gradient descent, Adagrad, and Adadelta. The generalization ability of the model is evaluated using cross-validation technology.

[0016] With the above technical solution, a deep learning model is used to provide fast and practical image processing for the image processing layer.

[0017] Preferably, it further includes an image management module, which includes an image acquisition unit and an image storage and management unit, used for classifying, annotating, and archiving medical images, providing annotation information for the intelligent diagnosis module to assist model training, and at the same time receiving the diagnosis results feedback by the intelligent diagnosis module as new annotation information for storage.

[0018] With the above technical solution, the image management module is constructed through the cooperation of the image acquisition unit and the image storage and management unit, which is convenient for...

[0019] Preferably, it further includes a user management and permission control module, used for managing system user information, setting the permission levels of different users, and ensuring the security of system data and the legality of access.

[0020] With the above technical solution, by setting user permissions, hierarchical display of system data is further provided, which is convenient for protecting the medical image data in the system.

[0021] Preferably, it further includes a data sharing and exchange module, which is connected to the user management and permission control module. Under the dual requirements of user account security and user permissions, it is associated with relevant information systems in the hospital, used for sharing and exchanging medical image data, and the information transmission of this module supports multiple secure network transmission protocols.

[0022] With the above technical solution, through the user account security status and user permission restrictions, fast query of various medical images is realized.

[0023] Preferably, it further includes an image data analysis module. The image data analysis module is connected to the image management module to obtain medical image data, the image data analysis module is connected to the intelligent diagnosis module to obtain diagnosis result data, and data mining and analysis technologies are set in the image data analysis module.

[0024] With the above technical solution, through further analysis of the image data, it assists doctors in diagnosing diseases.

[0025] Compared with the prior art, the beneficial effects of the present invention are: This intelligent management system for medical image information: 1. In the image preprocessing module, the image enhancement sub-module improves the overall contrast and detail clarity of the image through histogram equalization and contrast-limited adaptive histogram equalization techniques, enabling doctors to observe the lesion features in the image more clearly. The noise reduction sub-module uses Gaussian filtering and median filtering algorithms to effectively remove image noise and improve image quality, providing a more reliable image basis for subsequent diagnosis. Moreover, the image segmentation sub-module can accurately segment the abnormal regions in medical images with the help of deep learning algorithms, providing more accurate image data for medical diagnosis; And the application layer provides an intuitive and convenient operation interface for medical staff, supporting querying, browsing, comparative analysis, and various interactive operations of medical images, enabling medical staff to conveniently view and analyze images, improving work efficiency, and reducing human errors caused by complex operations. Furthermore, by connecting the hospital information system and the electronic medical record system through this system, the interconnection of data such as patient information and diagnosis results is realized, enabling doctors to directly obtain relevant medical images and diagnosis results while viewing patient medical records, comprehensively understanding the patient's condition, and providing strong support for formulating more accurate treatment plans; 2. The data acquisition layer can be seamlessly connected to various medical imaging devices through the standardized DICOM protocol to realize real-time acquisition of medical images, thereby improving the efficiency of image acquisition and reducing medical process delays caused by device incompatibility or acquisition delays; The image management module classifies, annotates, and archives medical images, facilitating image storage and retrieval. It also provides rich annotation information for the intelligent diagnosis module, helping to improve the accuracy of the diagnosis model. At the same time, the diagnosis results feedback by the intelligent diagnosis module are stored as new annotation information to further improve the annotation content of the images, forming a virtuous information cycle. Furthermore, based on the multi-dimensional indexing mechanism, the image retrieval and query module enables medical staff to quickly locate the required medical image data when querying images at the application layer, effectively saving the time for finding images, improving the efficiency of medical work, and enabling doctors to quickly obtain the patient's historical imaging data for comparative analysis. Brief Description of the Drawings

[0026] Figure 1 It is a schematic diagram of the module association of the intelligent management system for medical image information of the present invention; Figure 2 It is a schematic diagram of the architecture of the intelligent management system for medical image information of the present invention. Detailed Embodiments

[0027] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0028] Please refer to Figure 1 - Figure 2 , the present invention provides a technical solution: an intelligent management system for medical image information, including a data acquisition layer, a data storage layer, a data processing layer, and an application layer, wherein: The data acquisition layer includes an image acquisition unit, which is used to connect to various medical imaging devices through a standardized interface protocol, collect medical image data in real time, and has a data block transmission and caching function to ensure the stability and efficiency of data acquisition; The standardized interface protocol in the data acquisition layer is the DICOM protocol, which is compatible with a variety of medical imaging devices and standardizes data transmission. The data items in the data acquisition layer are transported to the data processing layer; The data storage layer has a distributed storage architecture, uses AES data encryption technology to encrypt and store medical image data, and is also equipped with a key management system to ensure key security. It also includes an image retrieval and query module for quickly retrieving and querying stored medical images; The distributed storage architecture is based on the Ceph distributed storage system and uses erasure coding technology to achieve redundant data storage; The data processing layer is provided with an image preprocessing module for preprocessing the collected medical images, and an intelligent diagnosis module for performing intelligent diagnosis on the preprocessed images using a deep learning model. The image preprocessing module includes an image enhancement sub-module, a noise reduction sub-module, and an image segmentation sub-module. The image enhancement sub-module uses histogram equalization and contrast-limited adaptive histogram equalization technologies to enhance the image contrast. The noise reduction sub-module uses algorithms such as Gaussian filtering and median filtering to remove image noise. The image segmentation sub-module uses a deep learning algorithm to segment abnormal regions in medical images. The deep learning model is a medical image diagnosis model based on a convolutional neural network, and uses transfer learning technology to initialize model parameters, performs model training through optimization algorithms such as stochastic gradient descent, Adagrad, and Adadelta, and uses cross-validation technology to evaluate the generalization ability of the model; The application layer provides an operation interface for medical staff to implement functions such as querying, browsing, and comparative analysis of medical images, supports image browsing and interaction, displays the diagnostic results output by the data processing layer, and can be connected to the hospital information system and the electronic medical record system to achieve information interconnection; Combined with the Figure 1As shown, when the data acquisition layer is in use, the internal image acquisition unit in it establishes connections with various medical imaging devices through the standardized DICOM protocol; The DICOM protocol ensures that the system can seamlessly interface with devices such as CT, MRI, and X-ray machines from different manufacturers and of different models. Once the device completes image acquisition, the image acquisition unit will immediately respond and obtain the image data in real time. The acquired data is uploaded and enters the data acquisition unit, which will process the image data in blocks and gradually transmit it to the system through the network. At the same time, a buffer area is set up to temporarily store data when the network is unstable or the data transmission speed fluctuates, avoiding data loss or transmission interruption, and ensuring that the data can enter the system efficiently and completely. The acquired data will be immediately transmitted to the data processing layer to provide the original material for subsequent processing; The data storage layer adopts a distributed storage architecture. Based on the Ceph distributed storage system, it dispersedly stores medical image data on multiple storage nodes. This architecture features high reliability, high scalability, and high performance; Through the erasure code technology, with a 4+2 erasure code strategy, the data is divided into 4 data blocks and 2 parity blocks, which are stored on different nodes respectively. Even if some nodes fail, the data can be restored through the parity blocks, ensuring the integrity and reliability of the data; When the medical image is being transmitted and used, the AES data encryption technology is adopted to encrypt and store the medical image data to prevent data leakage. The key management system is responsible for generating, storing, and managing the encryption keys to ensure the security of the keys. When obtaining the medical image data, only through the correct key can the encrypted data be decrypted and read. At the same time, each medical image carries identifiers such as patient information, examination time, and image type. When medical staff conduct image queries at the application layer, this module can quickly locate the target image data stored on the distributed nodes and return it to the application layer; The data processing layer receives the image data from the data acquisition layer, where: Image enhancement sub-module: Using histogram equalization technology, it redistributes the gray values of the image to make the gray distribution of the image more uniform and enhance the overall contrast. Contrast Limited Adaptive Histogram Equalization enhances the contrast of the image in local areas, better highlighting the image details and making the features in the image easier to identify; Noise reduction sub-module: Applying the Gaussian filtering algorithm, by weighted averaging the image pixel points and their neighborhoods, it smooths the image and removes Gaussian noise. The median filtering algorithm replaces the gray value of the pixel point with the median of the gray values of the pixels in its neighborhood, effectively removing impulse noise such as salt-and-pepper noise and improving the image quality; Image segmentation sub-module: Using deep learning algorithms to segment abnormal regions in medical images. Adopting network structures such as U-Net, through the training of a large amount of labeled data, the model can accurately identify regions of interest such as diseased tissues and organs in the image, separate them from the background, and provide more accurate data for subsequent intelligent diagnosis; The above-mentioned medical image processing realizes the quantification of various features in medical images. The diseased area is separated from the background through image segmentation algorithms, and then mathematical morphology algorithms and geometric calculation methods are used to obtain various quantitative indicators of the disease, helping doctors more intuitively understand the characteristics of the disease; Subsequently, the data enters the intelligent diagnosis module. The intelligent diagnosis module constructs a medical image diagnosis model based on a convolutional neural network. The preprocessed image is input into the model, and the model outputs the diagnosis results of the type, location, and degree of the disease in the medical image, and outputs the results to the application layer. During the training process of this diagnosis model, transfer learning technology is adopted, and the model parameters pre-trained on a large-scale natural image dataset are used to initialize the diagnosis model, reducing the training time and data requirements. Through optimization algorithms such as stochastic gradient descent, Adagrad, and Adadelta, the parameters of the model are continuously adjusted to minimize the loss function of the model on the training set. At the same time, cross-validation technology is used to evaluate the generalization ability of the model, prevent overfitting, and improve the analysis results of the medical image diagnosis model for the acquired intelligent images; When using this system, functions such as querying, browsing, and comparative analysis of medical images are provided for medical staff through the operation interface. When searching for images, the required medical images can be quickly queried by inputting conditions such as patient information and examination time. When browsing images, interactive operations such as zooming in, zooming out, panning, and rotating the images are supported, facilitating the observation of image details; At the same time, this system is docked with the hospital information system and the electronic medical record system. Through data interaction, the interconnection of data such as patient information and diagnosis results is realized, facilitating medical staff to comprehensively understand the patient's condition. Then, the diagnosis results output by the data processing layer are received and presented to the doctor in an intuitive way. During this period, the diseased area is marked on the image, and a text description is attached at the same time, providing assistance for the doctor's diagnosis; It also includes an image management module, which includes an image acquisition unit and an image storage and management unit, used to classify, label, and archive medical images, provide annotation information for the intelligent diagnosis module to assist model training, and at the same time receive the diagnosis results feedback by the intelligent diagnosis module as new annotation information for storage; Combined with the accompanying drawings of the specification Figure 2As shown, the image acquisition unit of the image management module is responsible for interacting with medical imaging equipment, collecting image data, and classifying, annotating and archiving the collected images. The classification is carried out according to dimensions such as patients, examination sites, and disease types; Annotation records key information in the image in detail, such as lesion characteristics, measurement data, etc. The images annotated in the image management module provide annotation information for the intelligent diagnosis module to help train the deep learning model and improve the accuracy of diagnosis. At the same time, the intelligent diagnosis module receives the diagnostic results fed back and stores them as new annotation information to continuously improve the annotation content of the image. It also includes a user management and authority control module, which is used to manage system user information, set authority levels for different users, and ensure the security of system data and the legitimacy of access; In conjunction with the accompanying drawings Figure 2 As shown, The user management and authority control module centrally manages all user information of the system, including user registration, login information, etc., and limits user authority when using the system. The authority setting is based on the user's role and different authority levels are set; Doctors can view and diagnose images, nurses can only view part of the image information, and administrators have the highest management authority in the system; Through strict permission control, the security of system data and the legitimacy of access are ensured to prevent unauthorized personnel from accessing sensitive data; It also includes a data sharing and exchange module, which is connected to the user management and authority control module, and is associated with the relevant information system in the hospital under the dual requirements of user account security and user authority, for sharing and exchanging medical image data, and the information transmission of this module supports a variety of secure network transmission protocols; In conjunction with the accompanying drawings Figure 2 As shown, when the data sharing and exchange module is used, it is closely connected with the user management and authority control module. Before data sharing and exchange, the user's account security and authority level are first verified. Only qualified users can initiate data sharing requests. Under the premise of meeting security and authority requirements, it is associated with relevant information systems in the hospital to achieve sharing and exchange of medical image data. By supporting multiple secure network transmission protocols, the security and integrity of data during transmission are ensured. Then, the image data in this system can be shared with other departments or hospitals, and image data from other systems can also be received, promoting the sharing and collaboration of medical resources. It also includes an image data analysis module, and the image data analysis module is connected to the image management module to obtain medical image data, the image data analysis module is connected to the intelligent diagnosis module to obtain diagnosis result data, and the image data analysis module is provided with data mining and analysis technology; in combination with the accompanying drawings of the specification Figure 2 As shown, by connecting with the image management module and the intelligent diagnosis module, a large amount of medical image data and diagnostic result data are obtained. These data form the basis of data analysis. By using data mining and analysis techniques, the obtained data are deeply mined; Through association analysis, the potential relationships between different diseases and image features are discovered. Through cluster analysis, the types and severities of diseases are classified and summarized. The analysis results can be fed back to the application layer to provide data support for medical research and clinical decision-making. At the same time, they can also help optimize the classification rules of the image management module and the algorithm model of the intelligent diagnosis module, improving the overall performance of the system.

[0029] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

Claims

1. An intelligent management system for medical image information, characterized in that, It includes a data acquisition layer, a data storage layer, a data processing layer, and an application layer, where: The data acquisition layer includes an image acquisition unit, which is used to connect to various medical imaging devices through a standardized interface protocol, collect medical image data in real time, and has functions of data block transmission and caching to ensure the stability and efficiency of data acquisition; The data storage layer has a distributed storage architecture, uses AES data encryption technology to encrypt and store medical image data, and is equipped with a key management system to ensure key security. It also includes an image retrieval and query module for quickly retrieving and querying stored medical images; The data processing layer is provided with an image preprocessing module for preprocessing the collected medical images, and an intelligent diagnosis module for performing intelligent diagnosis on the preprocessed images using a deep learning model; The application layer provides an operation interface for medical staff, realizes functions such as querying, browsing, and comparative analysis of medical images, supports image browsing and interaction, displays the diagnosis results output by the data processing layer, and can interface with the hospital information system and the electronic medical record system to achieve information interconnection.

2. The intelligent management system for medical image information according to claim 1, characterized in that: The standardized interface protocol in the data acquisition layer is the DICOM protocol, which is compatible with a variety of medical imaging devices and standardizes data transmission. The data items in the data acquisition layer are conveyed to the data processing layer.

3. An intelligent management system for medical image information according to claim 1, characterized in that: The distributed storage architecture is based on the Ceph distributed storage system and uses erasure coding technology to achieve redundant data storage.

4. An intelligent management system for medical image information according to claim 1, characterized in that: The image preprocessing module includes an image enhancement sub-module, a noise reduction sub-module, and an image segmentation sub-module. The image enhancement sub-module uses histogram equalization and contrast-limited adaptive histogram equalization technologies to enhance the image contrast.

5. An intelligent management system for medical image information according to claim 4, characterized in that: The noise reduction sub-module uses algorithms such as Gaussian filtering and median filtering to remove image noise. The image segmentation sub-module uses deep learning algorithms to segment abnormal regions in medical images.

6. The intelligent management system for medical image information according to claim 1, characterized in that: The deep learning model is a medical image diagnosis model constructed based on a convolutional neural network. It uses transfer learning technology to initialize model parameters, performs model training through optimization algorithms such as stochastic gradient descent, Adagrad, and Adadelta, and uses cross-validation technology to evaluate the generalization ability of the model.

7. An intelligent management system for medical image information according to claim 1, wherein: It also includes an image management module, which contains an image acquisition unit and an image storage and management unit, used to classify, label, and archive medical images, provide annotation information for the intelligent diagnosis module to assist model training, and at the same time receive the diagnosis results feedback by the intelligent diagnosis module as new annotation information for storage.

8. An intelligent management system for medical image information according to claim 1, characterized in that: It also includes a user management and permission control module for managing system user information, setting the permission levels of different users, and ensuring the security of system data and the legality of access.

9. An intelligent management system for medical image information according to claim 1, characterized in that: It also includes a data sharing and exchange module, which is connected to the user management and permission control module. Under the dual requirements of user account security and user permissions, it is associated with relevant information systems within the hospital for sharing and exchanging medical image data, and the information transmission of this module supports multiple secure network transmission protocols.

10. An intelligent management system for medical image information according to claim 1, characterized in that: It further includes an image data analysis module. The image data analysis module is connected to the image management module to obtain medical image data. The image data analysis module is connected to the intelligent diagnosis module to obtain diagnostic result data, and data mining and analysis technologies are set in the image data analysis module.

Images