Intelligent medical management system for bone injury patient

Abstract

This application discloses an intelligent medical management system for patients with bone injuries, including the following modules: a scanning module for acquiring imaging images of the affected area of ​​bone injury patients, including CT images, X-ray images, and ultrasound images; adjustable scanning parameters to adapt to the scanning needs of bone injuries in different locations and at different stages of the disease; and an image analysis module that automatically performs denoising, enhancement, and precise segmentation processing on the original scan images to achieve accurate quantitative assessment of the bone injury condition, generate standardized processed images and feature parameter reports, effectively reduce the workload of medical staff, improve diagnostic and treatment efficiency, and provide objective data support for clinical diagnosis and treatment. The prediction module calls historical case data and current patient disease progress data from the data storage module to identify high-risk groups for complications in advance and output targeted intervention suggestions, effectively avoiding poor patient rehabilitation, secondary injury, and aggravation of complications, thereby improving the quality of patient rehabilitation.

Description

Technical Field

[0001] This application relates to the technical field of medical management, and in particular to an intelligent medical management system for patients with bone injuries. Background Technology

[0002] Orthopedic injuries are common clinical conditions, encompassing various types such as fractures, bone contusions, and dislocations. Their diagnosis, treatment, and rehabilitation processes are characterized by long cycles, numerous stages, and significant individual differences. With the continuous development of medical technology, the traditional orthopedic medical management model is gradually transforming towards digitalization and intelligence, and various auxiliary management systems are being applied in clinical practice. However, existing technologies still have many shortcomings and are difficult to meet the needs of orthopedic patients for efficient medical management.

[0003] First, current technologies largely rely on radiologists manually interpreting images, visually assessing fracture lines, displacement, and callus formation. This is not only labor-intensive and inefficient, but also significantly influenced by the doctor's experience and subjective judgment, resulting in substantial individual differences. Even if some systems have basic image processing capabilities, they can only perform basic noise reduction, making it difficult to provide reliable data support for clinical diagnosis and treatment. This problem is particularly pronounced in primary healthcare institutions due to a lack of professional image analysis personnel.

[0004] Furthermore, current methods mainly rely on the clinical experience of medical staff to predict the fracture healing period, the probability of complications, and the rehabilitation effect. This is highly subjective. Although some systems have attempted to introduce statistical data methods, they cannot uncover the intrinsic relationship between disease progression data and prognostic results. It is difficult to identify high-risk groups for complications in advance, and it is also impossible to provide targeted intervention suggestions. This leads to poor rehabilitation effects for patients, and even problems such as secondary injury and aggravation of complications.

[0005] Application content

[0006] This application aims to address, at least to some extent, the technical problems in the related art.

[0007] To achieve the above objectives, this application proposes an intelligent medical management system for patients with bone injuries, comprising the following modules: a scanning module, used to acquire imaging images of the affected area of ​​patients with bone injuries, including CT images, X-ray images and ultrasound images, and adjustable scanning parameters to adapt to the scanning needs of bone injuries in different locations and at different stages of the disease, and outputting raw scan image data.

[0008] The bone injury image analysis module receives the raw scan image data output by the scanning module, performs denoising, enhancement, and segmentation processing on the image, extracts feature parameters of the bone injury site, including fracture line length, displacement distance, and callus thickness, and generates standardized processed images and feature parameter reports.

[0009] The patient condition management module is used to collect basic patient information, medical records, rehabilitation training data, and symptom feedback information. Combined with the feature parameters output by the bone injury image analysis module, it establishes a personal patient medical record and updates the disease progress data in real time.

[0010] The prediction module calls upon historical case data and current patient disease progress data from the data storage module, constructs a prediction model through machine learning algorithms, and predicts the patient's fracture healing period, probability of complications, and rehabilitation effect, outputting prediction results and intervention suggestions.

[0011] The data storage module is used to store raw scan images, processed images, feature parameter reports, patient medical records, historical case data, and prediction results. It uses encrypted storage to ensure data security and supports data querying, updating, and backup.

[0012] The interaction module enables two-way interaction between medical staff and the system, and between patients and the system. Medical staff can use the interaction module to set scanning parameters, view processing results, and modify treatment plans. Patients can use the interaction module to submit symptom feedback, view disease progress, and access rehabilitation guidance information.

[0013] In addition, the application may also include the following additional technical features:

[0014] Specifically, the scanning module includes a movable scanning terminal and a parameter adjustment unit. The parameter adjustment unit has multiple preset sets of adaptive parameter templates, which correspond to different bone injury sites such as limbs, trunk, and skull, as well as different disease stages such as acute phase, recovery phase, and healing phase. Medical staff can fine-tune the tube voltage, tube current, scanning slice thickness, and scanning speed based on the preset templates through the parameter adjustment unit.

[0015] Specifically, in the bone injury image analysis module, image denoising uses an algorithm combining Gaussian filtering and median filtering to remove electronic noise and environmental noise generated during the scanning process; image enhancement uses a histogram equalization algorithm to improve the contrast between the bone injury site and surrounding tissues; image segmentation uses a threshold segmentation method combined with an edge detection algorithm to accurately segment the fracture area, callus area, and surrounding soft tissue area. The feature parameter report includes fracture type, callus growth uniformity, and soft tissue swelling degree.

[0016] Specifically, the diagnosis and treatment records of the disease management module include diagnostic conclusions, medication lists, surgical records, follow-up examination times, and medical orders; rehabilitation training data includes training items, training duration, training frequency, and training completion rate; symptom feedback information allows patients to submit it in the form of text, voice, or images, and the system automatically extracts keywords from the feedback content and updates it synchronously to the patient's personal medical record, which supports displaying disease progression changes on a timeline.

[0017] Specifically, the prediction module uses a combination of random forest algorithm and BP neural network algorithm to construct the prediction model. The historical case data is cleaned and normalized, and cases that match the current patient's bone injury site, fracture type, age, and disease course are selected as training samples. The prediction results are output in the form of quantitative values ​​and levels. The probability of complication is divided into three levels: low, medium, and high. Intervention suggestions provide specific medication adjustments, rehabilitation training optimizations, and follow-up examination frequency suggestions for different risk levels.

[0018] Specifically, the data storage module uses an encryption algorithm to encrypt all data, and patient privacy information is stored after being desensitized; data backup adopts a combination of local backup and cloud backup, and supports fast querying by multiple dimensions such as patient ID, treatment time, and image type. The query operation requires verification of user identity and permissions.

[0019] Specifically, the interactive module is divided into a medical staff end and a patient end. The medical staff end has functions such as preset scanning parameters, image comparison and analysis, treatment plan editing, and batch management of patient files. The patient end has functions such as symptom feedback, disease progress query, viewing of rehabilitation training videos, medication reminders, and online consultation with medical staff. The system can automatically push personalized rehabilitation guidance information and follow-up reminders according to the patient's disease progress.

[0020] Specifically, the bone injury image analysis module also has an image comparison function, which can overlay and compare the currently processed standardized image with historical processed images, automatically mark the changes in fracture line, changes in callus growth and the subsidence of soft tissue swelling, and generate a comparison analysis report to assist medical staff in judging the progress of the disease.

[0021] In summary, the beneficial effects of the intelligent medical management system for bone injury patients proposed in this application are as follows:

[0022] 1. This application can automatically perform noise reduction, enhancement, and precise segmentation processing on the original scanned images through the bone injury image analysis module, without relying on medical staff to manually interpret the images. It can automatically extract feature parameters such as fracture line length, displacement distance, and callus growth thickness, realize accurate quantitative assessment of bone injury conditions, generate standardized processed images and feature parameter reports, effectively reduce the workload of medical staff, improve diagnostic and treatment efficiency, and provide objective data support for clinical diagnosis and treatment.

[0023] 2. By calling historical case data and current patient disease progress data from the data storage module through the prediction module, and using machine learning algorithms to build a prediction model, it can accurately predict the patient's fracture healing cycle, probability of complications, and rehabilitation effect. It can deeply explore the intrinsic relationship between disease progress data and prognosis, identify high-risk groups for complications in advance, and output targeted intervention suggestions, effectively avoiding poor patient rehabilitation, secondary injury, and aggravation of complications, thereby improving the quality of patient rehabilitation. Attached Figure Description

[0024] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0025] Figure 1 This is a flowchart of the modules of an intelligent medical management system for patients with bone injuries according to this application;

[0026] Figure 2 This application presents a flowchart of bone injury image acquisition and analysis for an intelligent medical management system for bone injury patients.

[0027] Figure 3 This application presents a flowchart of the disease management prediction and data interaction process for an intelligent medical management system for patients with bone injuries. Detailed Implementation

[0028] To make the technical means, inventive features, objectives, and effects of this application easier to understand, the application is further described below with reference to specific illustrations. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0029] The present application will now be described in further detail with reference to the accompanying drawings.

[0030] like Figures 1-3 As shown in this embodiment, an intelligent medical management system for patients with bone injuries includes the following modules:

[0031] The scanning module is used to acquire imaging images of the affected area of ​​patients with bone injuries, including CT images, X-ray images and ultrasound images. The scanning parameters can be adjusted to adapt to the scanning needs of bone injuries of different locations and different stages of the disease, and output raw scan image data.

[0032] It should be noted that this module utilizes the principles of multi-source image fusion and adaptive parameter adjustment. The system has a built-in multimodal scanning control unit with a parameter mapping library based on human anatomical structural features. When a scanning command is received for a specific bone injury site (such as the femur, spine, or phalanges), the control unit will automatically adjust the X-ray tube voltage (to adapt to X-ray imaging), CT slice thickness, and ultrasound probe frequency (to adapt to early soft tissue and callus imaging) according to the preset site-parameter matching matrix.

[0033] Through a closed-loop feedback control mechanism, the system monitors the image signal-to-noise ratio and sharpness in real time during the scanning process, and dynamically fine-tunes the exposure time or scanning step to ensure that the best coverage and imaging quality raw data can be obtained under different disease stages (such as acute fresh fractures and old callus growth period).

[0034] The bone injury image analysis module receives the raw scan image data output by the scanning module, performs denoising, enhancement, and segmentation processing on the image, extracts feature parameters of the bone injury site, including fracture line length, displacement distance, and callus thickness, and generates standardized processed images and feature parameter reports.

[0035] It should be noted that the deep learning semantic segmentation and multi-feature fusion algorithm first uses a neural network model based on the Transformer architecture to denoise and enhance the original scanned images to suppress imaging artifacts. Then, the fracture line, cortical bone, and callus region are accurately delineated using instance segmentation technology. Based on this, the spatial length and displacement distance of the fracture line are calculated using geometric morphology algorithms. The dynamic changes in callus growth thickness are analyzed by comparing time-series images. The quantitative geometric parameters are combined with qualitative image grading standards to generate a visualized standardized processed image and a structured feature parameter report.

[0036] It has enabled the transformation of bone injury conditions from subjective description to objective data. Traditionally, doctors relied on visual judgment to determine indicators such as displacement distance and callus thickness. Through algorithms, precise measurements at the micrometer level have been achieved, eliminating human visual errors and providing robust data support for efficacy evaluation. This has significantly shortened the time radiologists spend reading images.

[0037] The patient condition management module is used to collect basic patient information, medical records, rehabilitation training data, and symptom feedback information. Combined with the feature parameters output by the bone injury image analysis module, it establishes a personal patient medical record and updates the disease progress data in real time.

[0038] It should be noted that this module is based on medical information integration and data spatiotemporal fusion technology, which gathers multidimensional data from the registration system, electronic medical record system, scanning module, bone injury image analysis module and patient end. It uses graph database to build a disease process network with the patient's unique identifier as the core, and stores imaging features, clinical diagnosis and treatment records, rehabilitation training actions and symptom feedback in association. When the data of any module is updated, the system automatically triggers the retrospective and incremental update of the disease process file, forming a complete data closed loop of diagnosis-treatment-rehabilitation.

[0039] It not only enables seamless data flow throughout the entire process from emergency care and surgery to postoperative rehabilitation, allowing doctors to access all images and medical records of patients from the initial injury stage to the rehabilitation period for longitudinal comparative analysis, but also enables strong correlation analysis between image feature parameters and clinical symptoms and training data. For example, it can intuitively demonstrate the correlation between callus growth thickness and patient pain scores and joint mobility, providing a scientific basis for adjusting individualized treatment plans.

[0040] The prediction module calls upon historical case data and current patient disease progress data from the data storage module, constructs a prediction model through machine learning algorithms, and predicts the patient's fracture healing period, probability of complications, and rehabilitation effect, outputting prediction results and intervention suggestions.

[0041] It should be noted that, relying on machine learning time series prediction and big data evidence-based medicine technology, the data storage module pre-trains a deep prediction model based on massive historical case data. When the current patient's disease progress data (including the latest imaging features, physiological indicators, underlying diseases, etc.) is input, the model extracts key prognostic factors through feature engineering and combines time series analysis to perform regression prediction on the fracture healing cycle and classify and predict the probability of complications. At the same time, the model uses counterfactual reasoning algorithms to simulate the impact of different intervention methods (such as different rehabilitation programs) on the prognosis and outputs the optimal suggestions.

[0042] This module's AI model provides probabilistic and personalized predictions based on massive amounts of data. Doctors can use these predictions to develop intervention measures for high-risk individuals in advance, achieving preventative healthcare. The system not only predicts risks but also provides solutions. Through algorithms, it recommends rehabilitation training plans or drug interventions best suited to the patient's physical condition and healing stage, enhancing the scientific rigor and effectiveness of rehabilitation treatment.

[0043] The data storage module is used to store raw scan images, processed images, feature parameter reports, patient medical records, historical case data, and prediction results. It uses encrypted storage to ensure data security and supports data querying, updating, and backup.

[0044] It should be noted that a layered encryption and distributed storage architecture is employed. For sensitive data such as original images and medical records, end-to-end encrypted storage is implemented, and fine-grained access control is achieved based on attribute-based encryption technology. A hot / cold data separation strategy is adopted, with frequently accessed real-time data stored in a high-performance SSD array, while low-frequency historical case data is archived in low-cost distributed object storage. Simultaneously, data integrity and recoverability are ensured through scheduled snapshots and off-site disaster recovery mechanisms.

[0045] This module strictly adheres to medical data security regulations, employs multiple encryption technologies to comprehensively protect patient privacy from leakage, eliminates the risk of data tampering, and features a tiered storage architecture that balances access speed and cost, enabling rapid reading and writing of massive amounts of high-resolution image data. Furthermore, the system boasts excellent elastic scalability, allowing for seamless expansion as hospital workload grows.

[0046] The interaction module enables two-way interaction between medical staff and the system, and between patients and the system. Medical staff can use the interaction module to set scanning parameters, view processing results, and modify treatment plans. Patients can use the interaction module to submit symptom feedback, view disease progress, and access rehabilitation guidance information.

[0047] It should be noted that this module is based on a microservice architecture and a front-end / back-end separation Web / APP interaction design. The system connects to the medical staff management end and the patient user end through a unified API gateway. The medical staff end integrates complex parameter configuration, data visualization, and treatment plan editing functions; the patient end is lightweight, focusing on information display and feedback collection. The two-way interaction channel uses WebSocket long connection technology to ensure the real-time update of medical data and the issuance of instructions.

[0048] The interface is optimized for different user scenarios, the operation logic of the medical staff side conforms to the clinical workflow, reducing the learning cost, and the patient side interface is intuitive and easy to understand, improving user compliance.

[0049] In one embodiment of this application, the scanning module includes a movable scanning terminal and a parameter adjustment unit. The parameter adjustment unit presets multiple sets of adaptive parameter templates, which correspond to different bone injury sites such as limbs, trunk, and skull, as well as different disease stages such as acute phase, recovery phase, and healing phase. Medical staff can fine-tune the tube voltage, tube current, scanning slice thickness, and scanning speed based on the preset templates through the parameter adjustment unit.

[0050] Specifically, the mobile scanning terminal adopts a portable design, supports wired and wireless dual-mode connection, and can be flexibly moved to different medical scenarios such as wards and emergency rooms. It is suitable for the scanning needs of various bone injury patients, such as bedridden patients and critically ill patients in the emergency department. Patients do not need to move to a fixed scanning area, which greatly improves the efficiency of diagnosis and treatment and patient comfort.

[0051] The parameter adjustment unit has multiple preset sets of adaptive parameter templates. These templates are all optimized and generated based on a large amount of clinical bone injury diagnosis and treatment data. They are precisely corresponding to different bone injury sites such as the limbs (upper limbs, lower limbs), trunk (thoracic vertebrae, lumbar vertebrae, pelvis), and skull, as well as different disease stages such as the acute phase (early fracture stage, obvious swelling), recovery phase (initial callus formation), and healing phase (mature callus, blurred fracture line).

[0052] Medical staff can use the touch interface of the parameter adjustment unit to fine-tune the tube voltage (adjustment range of 40-120kV), tube current (adjustment range of 10-500mA), scanning slice thickness (adjustment range of 0.5-5mm), and scanning speed (adjustment range of 1-10 frames / second) based on the corresponding preset template and the patient's specific situation (such as age, physical condition, and severity of fracture). This ensures that the scanned images can clearly present the details of the bone injury while minimizing the radiation dose, thus balancing diagnostic accuracy and patient safety.

[0053] In one embodiment of this application, the bone injury image analysis module employs an algorithm combining Gaussian filtering and median filtering for image denoising to remove electronic noise and environmental noise generated during the scanning process; image enhancement uses a histogram equalization algorithm to improve the contrast between the bone injury site and surrounding tissues; image segmentation uses a threshold segmentation method combined with an edge detection algorithm to accurately segment the fracture area, callus area, and surrounding soft tissue area, and the feature parameter report includes fracture type, callus growth uniformity, and soft tissue swelling degree.

[0054] Specifically, the image denoising process employs a hybrid algorithm combining Gaussian filtering and median filtering. Gaussian filtering primarily suppresses electronic noise generated during the scanning process (such as detector thermal noise and circuit noise). By performing weighted smoothing on image pixels, it reduces the interference of noise on image details. Median filtering focuses on removing environmental noise (such as light interference and noise generated by equipment vibration). By replacing the original pixel with the median value within the pixel's neighborhood, it effectively preserves key details such as fracture edges and avoids image blurring during the filtering process.

[0055] The image enhancement process employs a histogram equalization algorithm, which adjusts the grayscale distribution of the image and stretches the dynamic range of grayscale, significantly improving the grayscale contrast between the bone injury site (such as fracture line and callus) and the surrounding normal tissue. This makes the originally blurry bone injury details clearer and easier for medical staff to quickly locate the lesion area.

[0056] The image segmentation process employs a threshold segmentation method combined with an edge detection algorithm. First, the threshold segmentation method is used to preliminarily divide the suspected lesion area into normal tissue areas. Then, the edge detection algorithm is used to accurately extract the area boundaries, thereby achieving precise segmentation of the fracture area, callus area, and surrounding soft tissue area. The feature parameter report includes detailed information such as fracture type (e.g., closed fracture, open fracture, comminuted fracture, linear fracture), callus growth uniformity (divided into three levels: uniform, relatively uniform, and non-uniform), and soft tissue swelling degree (divided into three levels: mild, moderate, and severe), providing quantitative basis for disease diagnosis.

[0057] In one embodiment of this application, the diagnosis and treatment records of the disease management module include diagnostic conclusions, medication lists, surgical records, follow-up examination times, and medical orders; rehabilitation training data includes training items, training duration, training frequency, and training completion rate; symptom feedback information allows patients to submit it in the form of text, voice, or images, and the system automatically extracts keywords from the feedback content and updates it synchronously to the patient's personal medical record, which supports displaying disease progression changes on a timeline.

[0058] Specifically, rehabilitation training data is updated in real time, covering training items (such as joint range of motion training, muscle strength training, balance training, etc.), training duration (the specific number of minutes for each training session), training frequency (number of training sessions per day / week), and training completion rate (the percentage of training actually completed by the patient compared to the preset training), which makes it easier for medical staff to keep track of the patient's rehabilitation status in real time.

[0059] Symptom feedback can be submitted by patients in three formats: text, voice, or image. The system incorporates natural language processing and image recognition algorithms to automatically extract keywords from the feedback (such as increased pain, persistent swelling, etc.) and simultaneously update the extracted key information to the patient's personal medical record. This record clearly displays the patient's entire disease progression from initial consultation, diagnosis, treatment to recovery along a timeline, including scanned images, medical records, recovery data, and symptom feedback at each time point, facilitating medical staff to track the disease progression and adjust treatment plans.

[0060] In one embodiment of this application, the prediction module uses a combination of random forest algorithm and BP neural network algorithm to construct a prediction model. The historical case data is cleaned and normalized, and cases that match the current patient's bone injury site, fracture type, age, and disease course are selected as training samples. The prediction results are output in the form of quantitative values ​​and levels. The probability of complication is divided into three levels: low, medium, and high. Intervention suggestions provide specific medication adjustments, rehabilitation training optimizations, and follow-up frequency suggestions for different risk levels.

[0061] Specifically, historical case data comes from the hospital's past bone injury diagnosis and treatment database. After data cleaning and normalization, a feature matching algorithm is used to select cases that highly match the current patient's bone injury location, fracture type, age, disease course, physical condition, and other parameters as training samples. This ensures the relevance and effectiveness of the training samples and improves the model's prediction accuracy.

[0062] The output of the prediction model is presented in both quantitative numerical and graded form. The probability of complication is expressed as a quantitative value of 0-100%, and is divided into three grades: low (0-30%), medium (31-70%), and high (71-100%), which makes it easier for medical staff to quickly judge the degree of risk.

[0063] Intervention recommendations provide specific and actionable guidance for different risk levels: for low-risk levels, it is recommended to maintain the current medication and rehabilitation training plan and have regular follow-up examinations; for medium-risk levels, it is recommended to slightly adjust the medication dosage, optimize the intensity of rehabilitation training, and appropriately increase the frequency of follow-up examinations; for high-risk levels, it is recommended to immediately adjust the medication plan, suspend high-intensity rehabilitation training, increase the number of follow-up examinations, and take further intervention and treatment measures if necessary to avoid the risk of complications in advance.

[0064] In one embodiment of this application, the data storage module uses an encryption algorithm to encrypt all data, and patient privacy information is stored after being desensitized; data backup adopts a combination of local backup and cloud backup, and supports fast querying by multiple dimensions such as patient ID, treatment time, and image type. The query operation requires verification of user identity and permissions.

[0065] Specifically, patient privacy information (such as name, ID number, contact information, home address, etc.) is stored after being anonymized, using methods such as hiding key fields and encrypting replacements. Data is backed up locally on the hospital server and automatically backed up regularly. Cloud backup uses encrypted cloud storage services to achieve off-site backup, effectively avoiding data loss due to local server failures, natural disasters, or other factors.

[0066] Meanwhile, system query operations require verification of user identity and permissions. Different roles (medical staff, administrators, and patients) are assigned different query permissions. Medical staff can only query the relevant data of the patients they are responsible for, patients can only query their own disease progress data, and administrators have full data query permissions to ensure the security and standardization of data access.

[0067] In one embodiment of this application, the interaction module is divided into a medical staff end and a patient end. The medical staff end has functions such as preset scanning parameters, image comparison and analysis, treatment plan editing, and batch management of patient files. The patient end has functions such as symptom feedback, disease progress query, viewing rehabilitation training videos, medication reminders, and online consultation with medical staff. The system can automatically push personalized rehabilitation guidance information and follow-up reminders according to the patient's disease progress.

[0068] Specifically, the medical staff's scanning parameter preset can save commonly used parameter templates, eliminating the need for repeated adjustments with each scan. The image comparison analysis can quickly retrieve and compare scan images of the patient at different times to help determine changes in the condition. The treatment plan editing allows medical staff to modify diagnostic conclusions, medication lists, rehabilitation plans, and other content in real time according to the patient's condition, and push them to the patient's end simultaneously.

[0069] Patients can log in conveniently via mobile app, WeChat mini-program, etc. Symptom feedback allows them to submit their discomfort symptoms at any time, which are quickly synchronized to the medical staff. Patients can view their personal medical records, rehabilitation data, and predicted results at any time. Standardized rehabilitation training videos are provided to guide patients in completing the training correctly. Medication reminders allow patients to set reminders for medication time and dosage to avoid missed or incorrect doses. Online consultation with medical staff allows patients to consult with responsible medical staff about their condition and rehabilitation-related issues in real time, with timely responses improving the patient's medical experience.

[0070] In addition, the system can automatically push personalized rehabilitation guidance information (such as targeted training suggestions and dietary guidance) and follow-up reminders (1-2 days in advance via pop-up windows and SMS) based on the patient's disease progression (such as from the acute phase to the recovery phase), helping patients to carry out standardized rehabilitation.

[0071] In one embodiment of this application, the bone injury image analysis module also has an image comparison function, which can overlay and compare the currently processed standardized image with the historical processed image, automatically mark the changes in fracture line, changes in callus growth and the subsidence of soft tissue swelling, generate a comparison analysis report, and assist medical staff in judging the progress of the disease.

[0072] Specifically, the system assists medical staff in assessing disease progression by precisely overlaying and comparing the currently processed standardized images with the patient's historical processed images. It supports adjusting image transparency for clear observation of changes in the bone injury site. Simultaneously, the system automatically marks key changes using image recognition algorithms, including changes in the fracture line, callus growth, and the resolution of soft tissue swelling. The markings are clear and intuitive, and the position and description of the markings can be manually adjusted.

[0073] After the comparison is completed, the system automatically generates a comparison analysis report, which includes comparison images, detailed descriptions of changes, judgments of change trends, and preliminary treatment suggestions. Medical staff can use this report to quickly assess the patient's recovery and adjust the treatment and rehabilitation plan in a timely manner, thereby improving the accuracy and timeliness of diagnosis and treatment.

[0074] It should be noted that, in this document, the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0075] The present application and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present application. The actual structure is not limited to this. In conclusion, if a person skilled in the art is inspired by this description and designs a similar structure and embodiment without departing from the spirit of the present application, such design should fall within the protection scope of the present application.

Claims

1. An intelligent medical management system for patients with bone injuries, characterized in that, Includes the following modules: The scanning module is used to acquire imaging images of the affected area of ​​patients with bone injuries, including CT images, X-ray images and ultrasound images. The scanning parameters can be adjusted to adapt to the scanning needs of bone injuries of different locations and different stages of the disease, and output raw scan image data. The bone injury image analysis module receives the raw scan image data output by the scanning module, performs denoising, enhancement, and segmentation processing on the image, extracts feature parameters of the bone injury site, including fracture line length, displacement distance, and callus thickness, and generates standardized processed images and feature parameter reports. The disease management module is used to collect basic patient information, medical records, rehabilitation training data and symptom feedback information. Combined with the feature parameters output by the bone injury image analysis module, it establishes a personal patient medical record and updates the disease progress data in real time. The prediction module calls upon historical case data and current patient disease progress data from the data storage module, constructs a prediction model through machine learning algorithms, and predicts the patient's fracture healing period, probability of complications and rehabilitation effect, outputting prediction results and intervention suggestions. The data storage module is used to store raw scan images, processed images, feature parameter reports, patient medical records, historical case data and prediction results. It uses encrypted storage to ensure data security and supports data query, update and backup. The interaction module enables two-way interaction between medical staff and the system, and between patients and the system. Medical staff can use the interaction module to set scanning parameters, view processing results, and modify treatment plans. Patients can use the interaction module to submit symptom feedback, view disease progress, and access rehabilitation guidance information.

2. The intelligent medical management system for patients with bone injuries according to claim 1, characterized in that, The scanning module includes a movable scanning terminal and a parameter adjustment unit. The parameter adjustment unit has multiple preset sets of adaptive parameter templates, which correspond to different bone injury sites such as limbs, trunk, and skull, as well as different disease stages such as acute phase, recovery phase, and healing phase. Medical staff can fine-tune the tube voltage, tube current, scanning slice thickness, and scanning speed based on the preset templates through the parameter adjustment unit.

3. The intelligent medical management system for patients with bone injuries according to claim 1, characterized in that, In the bone injury image analysis module, image denoising uses an algorithm combining Gaussian filtering and median filtering to remove electronic noise and environmental noise generated during the scanning process; image enhancement uses a histogram equalization algorithm to improve the contrast between the bone injury site and surrounding tissues; image segmentation uses a threshold segmentation method combined with an edge detection algorithm to accurately segment the fracture area, callus area, and surrounding soft tissue area. The feature parameter report includes fracture type, callus growth uniformity, and soft tissue swelling degree.

4. The intelligent medical management system for patients with bone injuries according to claim 1, characterized in that, The medical records in the disease management module include diagnostic conclusions, medication lists, surgical records, follow-up examination times, and medical orders; the rehabilitation training data includes training items, training duration, training frequency, and training completion rate. Symptom feedback information can be submitted by patients in the form of text, voice or images. The system automatically extracts keywords from the feedback content and updates it to the patient's personal medical record. The record supports displaying the changes in the course of the disease on a timeline.

5. The intelligent medical management system for patients with bone injuries according to claim 1, characterized in that, The prediction module uses a combination of random forest algorithm and BP neural network algorithm to construct the prediction model. The historical case data is cleaned and normalized, and cases that match the current patient's bone injury site, fracture type, age, and disease course are selected as training samples. The prediction results are output in the form of quantitative values ​​and levels. The probability of complication is divided into three levels: low, medium, and high. Intervention suggestions provide specific medication adjustments, rehabilitation training optimizations, and follow-up examination frequency suggestions for different risk levels.

6. The intelligent medical management system for patients with bone injuries according to claim 1, characterized in that, The data storage module uses encryption algorithms to encrypt all data, and patient privacy information is stored after being desensitized. Data backup adopts a combination of local backup and cloud backup, and supports fast querying by multiple dimensions such as patient ID, treatment time, and image type. The query operation requires verification of user identity and permissions.

7. The intelligent medical management system for patients with bone injuries according to claim 1, characterized in that, The interactive module is divided into a medical staff end and a patient end. The medical staff end has functions such as preset scanning parameters, image comparison and analysis, treatment plan editing, and batch management of patient files. The patient end has functions such as symptom feedback, disease progress query, viewing of rehabilitation training videos, medication reminders, and online consultation with medical staff. The system can automatically push personalized rehabilitation guidance information and follow-up reminders according to the patient's disease progress.

8. The intelligent medical management system for patients with bone injuries according to claim 2, characterized in that, The bone injury image analysis module also has an image comparison function, which can overlay and compare the currently processed standardized image with historical processed images, automatically mark the changes in fracture line, callus growth, and soft tissue swelling, and generate a comparison analysis report to assist medical staff in judging the progress of the disease.

Images