An artificial intelligence-based intelligent evaluation system after anterior cruciate ligament reconstruction

Abstract

The application discloses an artificial intelligence-based intelligent evaluation system for anterior cruciate ligament reconstruction, comprising a data acquisition module for collecting clinical information and MRI image data of a patient; a ligament graft maturity evaluation module for automatically measuring and calculating the maturity of the anterior cruciate ligament graft; and a tendon-bone interface healing evaluation module for automatically measuring the change of the bone tunnel diameter and the change of the edema around the tunnel. The application provides precise evaluation and personalized suggestions for the postoperative recovery of patients with anterior cruciate ligament reconstruction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical image processing, in particular to an anterior cruciate ligament reconstruction postoperative intelligent evaluation system based on artificial intelligence. BACKGROUND

[0002] The anterior cruciate ligament (ACL) is one of the most important ligaments in the knee joint, mainly responsible for limiting the forward movement of the tibia relative to the femur and rotational stability. ACL injury is very common in sports injuries, especially in high-intensity sports. ACL reconstruction (ACLR) is the main treatment method to restore knee stability and function.

[0003] After ACL reconstruction, evaluating the maturity of the graft and the healing of the tendon-bone interface is crucial for guiding rehabilitation training. The graft undergoes a series of biological processes after surgery, including osseointegration and ligamentization, until its properties approach those of the native ligament. This process is divided into several stages: early stage (about 4 weeks), proliferation stage (4 to 12 weeks), and final ligamentization stage (more than 1 year). The maturity of the graft and the healing of the tendon-bone interface directly affect the development of rehabilitation plans and the return to sports. Early return to sports can increase the risk of graft failure, so rehabilitation plans must consider the biological properties of the graft at different stages to avoid excessive loading.

[0004] Traditional evaluation methods mainly rely on the experience of clinicians and manual measurements of imaging. This method has many limitations, including subjectivity of evaluation, complexity of operation, and insufficient accuracy of evaluation results, which can easily lead to misjudgment or delay of rehabilitation progress. This subjective evaluation method not only depends on the experience level of the doctor, but also is affected by human factors, leading to inaccurate guidance during postoperative rehabilitation. SUMMARY

[0005] To overcome the technical problems of strong subjectivity, complex operation, and lack of precise rehabilitation guidance in the evaluation after anterior cruciate ligament reconstruction, the present application provides an anterior cruciate ligament reconstruction postoperative intelligent evaluation system based on artificial intelligence, which provides more accurate and reliable postoperative evaluation results, thereby better guiding the rehabilitation process of patients, optimizing rehabilitation plans, and reducing the risk of graft failure.

[0006] To achieve the above-mentioned purpose, the present application provides an anterior cruciate ligament reconstruction postoperative intelligent evaluation system based on artificial intelligence, comprising a data acquisition module, a ligament graft maturity evaluation module, and a tendon-bone interface healing evaluation module.

[0007] The data acquisition module is used to collect the clinical information and MRI image data of the patient.

[0008] The ligament graft maturity assessment module is used to automatically measure and calculate the maturity of the anterior cruciate ligament graft;

[0009] The tendon-bone interface healing assessment module is used to automatically measure the changes of bone tunnel diameter and peritunnel edema;

[0010] The data acquisition module is connected with the ligament graft maturity assessment module and the tendon-bone interface healing assessment module respectively.

[0011] Preferably, the data acquisition module comprises:

[0012] The clinical information acquisition unit is used to retrieve the basic information and surgical information of the patient;

[0013] The MRI data acquisition unit is used to retrieve the follow-up knee MRI data of the patient.

[0014] Preferably, the ligament graft maturity assessment module comprises:

[0015] The ligament graft and background segmentation unit is used to segment the anterior cruciate ligament graft and background in the knee MRI data by an artificial intelligence automatic identification method;

[0016] The signal measurement unit is used to measure the signal intensity of the background and the anterior cruciate ligament graft respectively;

[0017] The signal-to-noise ratio calculation unit is used to calculate the signal-to-noise ratio of the anterior cruciate ligament graft at different positions, and assess the maturity of the anterior cruciate ligament graft based on the signal-to-noise ratio.

[0018] Preferably, the process of measuring the background signal by the signal measurement unit comprises:

[0019] Selecting the sagittal position of the knee MRI T2 fat-suppressed sequence;

[0020] Automatically selecting a circle with a first preset distance in diameter in the segmented background by artificial intelligence and measuring the signal intensity thereof as the background signal;

[0021] The method for measuring the signal intensity of the anterior cruciate ligament graft comprises:

[0022] Segmenting the anterior cruciate ligament graft based on a convolutional neural network to obtain the distal end, middle part and proximal end of the anterior cruciate ligament graft, and automatically selecting a circle with a first preset distance in diameter in the distal end, middle part and proximal end of the anterior cruciate ligament graft and measuring the signal intensity thereof.

[0023] Preferably, the process of calculating the signal-to-noise ratio of the anterior cruciate ligament graft at different positions by the signal-to-noise ratio calculation unit comprises:

[0024] The signal-to-noise ratio is obtained by subtracting the contrast signal from the signal of any of the distal end, middle part or proximal end of the anterior cruciate ligament graft and dividing by the background signal;

[0025] The contrast signal is measured by automatically selecting a circle with a first preset distance in diameter at a preset position and measuring the signal intensity, and the preset position includes the quadriceps tendon tissue and the posterior cruciate ligament tissue.

[0026] Preferably, the tendon-bone interface healing evaluation module comprises:

[0027] A bone tunnel change detection unit is configured to detect a change in the bone tunnel.

[0028] A peritunnel edema change detection unit is configured to detect a change in edema around the bone tunnel.

[0029] Preferably, the process of detecting the change in the bone tunnel by the bone tunnel change detection unit comprises:

[0030] A T2 fat-suppressed sequence axial view and a T2 fat-suppressed sequence sagittal view of the knee joint MRI are selected.

[0031] The entry level of the ligament graft into the distal femur and the proximal tibia is determined according to the T2 fat-suppressed sequence sagittal view, and the axial view of the femoral tunnel and the tibial tunnel is positioned.

[0032] The tunnel diameter is measured on the T2 axial view level of the femoral tunnel and the tibial tunnel based on a convolutional neural network, respectively.

[0033] The in-situ tunnel diameter is called by the data acquisition module, the ratio of the tunnel diameter to the in-situ tunnel diameter is measured, and the change in the bone tunnel is determined.

[0034] Preferably, the process of detecting the change in edema around the bone tunnel by the peritunnel edema change detection unit comprises:

[0035] In the T2 fat-suppressed sequence of the same level of the femoral tunnel and the tibial tunnel, a circle with a second preset distance in diameter is drawn with the center of the femoral tunnel and the tibial tunnel as the center, respectively, and the signal intensity is measured as the signal intensity around the bone tunnel.

[0036] A circle with the second preset distance in diameter is drawn in the middle part of the femur without bone edema, and the bone marrow signal intensity is measured as a reference value.

[0037] The change in edema around the bone tunnel is determined by the ratio of the signal around the bone tunnel to the reference value.

[0038] Preferably, the system further comprises a rehabilitation recommendation module, and the rehabilitation recommendation module comprises:

[0039] Rehabilitation database: for literature retrieval, research on the correlation between the maturity of the ligament graft after ACL reconstruction and rehabilitation training, rehabilitation methods and actions, and rehabilitation guidelines;

[0040] Early warning unit: for providing rehabilitation strategies for patients by the evaluation results of the maturity of the ligament graft and the healing of the tendon-bone interface, in combination with the rehabilitation database.

[0041] Preferably, the rehabilitation database can also optimize and update the database content in combination with the evaluation results and rehabilitation results of the patients.

[0042] Compared with the prior art, the present application has the following advantages and technical effects:

[0043] The present application objectively evaluates the maturity of the graft and the healing of the tendon-bone interface by automatic collection and analysis of patient data using a convolutional neural network model, which reduces the influence of human error and provides more accurate and reliable postoperative evaluation results, thereby better guiding the rehabilitation process of the patient, optimizing the rehabilitation plan and reducing the risk of graft failure. BRIEF DESCRIPTION OF DRAWINGS

[0044] The accompanying drawings, which form a part of this application, are intended to provide further understanding of the application and are incorporated herein in their entirety. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0045] Figure 1 A structure schematic diagram of an artificial intelligence-based intelligent evaluation system for ACL reconstruction according to an embodiment of the present application;

[0046] Figure 2 A schematic diagram of ligament graft maturity evaluation according to an embodiment of the present application;

[0047] Figure 3 A schematic diagram of tendon-bone healing evaluation according to an embodiment of the present application;

[0048] Figure 4 A convolutional neural network algorithm and data processing conceptual diagram according to an embodiment of the present application. DETAILED DESCRIPTION

[0049] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0050] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown.

[0051] The present application provides an artificial intelligence-based intelligent evaluation system for anterior cruciate ligament reconstruction, which comprises a data acquisition module, a ligament graft maturity evaluation module and a tendon-bone interface healing evaluation module. Figure 1 , including a data acquisition module, a ligament graft maturity evaluation module and a tendon-bone interface healing evaluation module;

[0052] The data acquisition module is used to collect the clinical information and MRI image data of the patient;

[0053] The ligament graft maturity evaluation module is used to automatically measure and calculate the maturity of the anterior cruciate ligament graft;

[0054] The tendon-bone interface healing evaluation module is used to automatically measure the change of bone tunnel diameter and the change of edema around the tunnel;

[0055] The data acquisition module is connected with the ligament graft maturity evaluation module and the tendon-bone interface healing evaluation module respectively.

[0056] The present application automatically collects and analyzes patient data, and objectively evaluates the maturity of the graft and the healing of the tendon-bone interface using a convolutional neural network model. This method reduces the influence of human error and provides more accurate and reliable postoperative evaluation results, thereby better guiding the rehabilitation process of the patient, optimizing the rehabilitation plan and reducing the risk of graft failure.

[0057] Further, the data acquisition module comprises:

[0058] Clinical information acquisition unit: used to retrieve patient basic information and surgical information;

[0059] MRI data acquisition unit: used to retrieve follow-up knee MRI data of the patient.

[0060] Specifically, the data acquisition module is used to retrieve patient basic information, surgical information and follow-up knee MRI data from a hospital case system and an image system;

[0061] In this embodiment, the knee MRI data includes but is not limited to T2 fat-suppressed sagittal and coronal, PD fat-suppressed sequence sagittal and coronal, etc.

[0062] Further, the ligament graft maturity evaluation module comprises:

[0063] Ligament graft and background segmentation unit: used to segment the anterior cruciate ligament graft and background in the knee joint MRI data using an artificial intelligence automatic identification method;

[0064] Signal measurement unit: used to measure the signal intensity of the background and the anterior cruciate ligament graft, respectively;

[0065] Signal-to-noise ratio calculation unit: used to calculate the signal-to-noise ratio of the anterior cruciate ligament graft at various sites, and to assess the maturity of the anterior cruciate ligament graft based on the signal-to-noise ratio.

[0066] Specifically, the method for measuring the background signal is as follows:

[0067] Select sagittal T2 fat-suppressed MRI sequence for the knee joint;

[0068] Using artificial intelligence, a circle with a diameter of a first preset distance is automatically selected from the segmented background and its signal strength is measured as the background signal.

[0069] In this embodiment, the first preset distance is 3mm, and the background signal is automatically selected by artificial intelligence with a diameter of 3mm and its signal strength is measured.

[0070] The method for measuring the signal intensity of anterior cruciate ligament grafts is as follows:

[0071] The anterior cruciate ligament (ACL) graft is segmented based on a convolutional neural network to obtain the distal, middle, and proximal ends of the ACL graft. Circles with a diameter of a first preset distance are automatically selected at the distal, middle, and proximal ends of the ACL graft, and their signal strength is measured.

[0072] Furthermore, the signal-to-noise ratio of the anterior cruciate ligament graft at different sites was calculated, such as... Figure 2 ,include:

[0073] The signal-to-noise ratio is obtained by subtracting the contrast signal from the signal at any of the distal, middle, or proximal parts of the anterior cruciate ligament graft and then dividing by the background signal.

[0074] Specifically, a circle with a diameter of a first preset distance is automatically selected at a preset location and the signal intensity is measured. The measurement result is used as the comparison signal. The preset location includes the quadriceps tendon tissue and the posterior cruciate ligament tissue.

[0075] Specifically, the anterior cruciate ligament graft signal is automatically identified and segmented using artificial intelligence, and a circle with a diameter of 3 mm is automatically selected at the distal, middle and proximal ends of the graft and its signal intensity (SI) is measured.

[0076] The comparison signal is automatically identified by artificial intelligence for the quadriceps tendon, and a circle with a diameter of 3mm is automatically selected at the distal end of the quadriceps tendon and its signal intensity is measured as the comparison signal; the comparison signal source includes, but is not limited to, normal tendon tissues such as the quadriceps tendon and the posterior cruciate ligament.

[0077] In this embodiment, sagittal images of the knee joint using a T2 fat-suppressed MRI sequence were used.

[0078] Select a layer that provides a complete view of the ligament graft and the distal quadriceps tendon; automatically segment the anterior cruciate ligament graft using a convolutional neural network, dividing it into distal, middle, and proximal sections, and labeling them with 3mm diameter circles as L1, L2, and L3 respectively; the convolutional neural network training process is as follows. Figure 4 As shown;

[0079] The distal end of the quadriceps tendon is automatically identified using a convolutional neural network and marked with a 3mm diameter circle, labeled as C1.

[0080] The background is identified using a convolutional neural network and labeled with a circle of 3mm diameter, marked as B1.

[0081] The signal measurement unit calculates the signal intensity (SI) within the marked areas.

[0082] The signal-to-noise ratio (SNR) of ligament grafts is calculated as follows (taking the distal end as an example): Distal ligament graft SNR = (Distal ligament graft SI – Distal quadriceps tendon SI) / Background SI.

[0083] In this embodiment, the artificial intelligence method for automatically detecting anterior cruciate ligament grafts, quadriceps tendons, and the background is based on convolutional neural networks, including but not limited to YOLO, U-Net, and other convolutional neural networks.

[0084] Furthermore, the tendon-bone interface healing assessment module includes:

[0085] Bone tunnel change detection unit: used to detect changes in bone tunnels;

[0086] Tunnel peri-tunnel edema change detection unit: used to detect edema changes around bone tunnels.

[0087] Specifically, detecting changes in bone tunnels, such as Figure 3 ,include:

[0088] Select axial and sagittal views of T2 fat-suppressed MRI sequences for the knee joint;

[0089] The entry level of the ligament graft in the distal femur and proximal tibia was determined based on the sagittal position of the T2 fat suppression sequence, and the axial position of the femoral tunnel and tibial tunnel was located.

[0090] The diameters of the femoral and tibial tunnels were measured on the T2 axial plane based on a convolutional neural network.

[0091] The intraoperative tunnel diameter is retrieved by the data acquisition module, and the ratio of the tunnel diameter to the intraoperative tunnel diameter is measured to determine the changes in the bone tunnel.

[0092] Detecting changes in edema around bone tunnels, including:

[0093] In the T2 fat suppression sequence at the same level as the femoral tunnel and the tibial tunnel, circles with a diameter of a second preset distance are drawn with the center of the femoral tunnel and the center of the tibial tunnel, respectively, and the signal intensity is measured and used as the signal around the bone tunnel.

[0094] Draw a circle with a diameter of the second preset distance in the non-osteoedema area of ​​the middle femur, measure the bone marrow signal intensity, and use the bone marrow signal intensity as a reference value;

[0095] Changes in bone tunnel edema are determined by the ratio of the signal around the bone tunnel to the reference value.

[0096] Specifically, in this embodiment, the second preset distance is 10mm.

[0097] This embodiment introduces a convolutional neural network to automate the assessment of ligament graft maturity and tendon-bone interface healing. Since the tasks involved in this project all utilize basic deep learning model training methods, no specific personalized explanations are provided. The model training method is as follows: first, the data is labeled; second, necessary image preprocessing is performed, including resizing and normalization; then, the model is trained using a selected convolutional neural network; and finally, the model's performance is evaluated using metrics such as mean absolute error and mean squared error.

[0098] Furthermore, the system also includes a rehabilitation suggestion module, which includes:

[0099] Rehabilitation database: used for literature retrieval, research on the relationship between the maturity of ligament grafts after anterior cruciate ligament reconstruction and rehabilitation training, rehabilitation methods and movements, and rehabilitation guidelines;

[0100] Early warning unit: Used to provide rehabilitation strategies for patients based on the assessment results of ligament graft maturity and tendon-bone interface healing, combined with the rehabilitation database.

[0101] The rehabilitation database can also optimize and update its content by combining patients' rehabilitation outcomes and assessment results.

[0102] Specifically, the workflow of the rehabilitation advice module is as follows:

[0103] First, in the process of building the rehabilitation database, a systematic literature search was conducted to conduct in-depth research on the relationship between the maturity of ligament grafts after anterior cruciate ligament reconstruction and rehabilitation training, current rehabilitation methods and movements, and domestically and internationally recognized rehabilitation guidelines.

[0104] The early warning unit provides personalized rehabilitation strategies for patients based on the assessment results of ligament graft maturity and tendon-bone interface healing, combined with the rehabilitation database.

[0105] The rehabilitation database can also be optimized and updated by combining patients' rehabilitation outcomes and assessment results.

[0106] This invention is based on artificial intelligence and uses convolutional neural networks to provide objective and accurate assessments of graft maturity and tendon-bone interface healing. It effectively develops rehabilitation plans, improves the accuracy and consistency of assessments, provides personalized rehabilitation advice, reduces the burden on doctors, and optimizes the rehabilitation process.

[0107] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An artificial intelligence-based intelligent evaluation system after anterior cruciate ligament reconstruction, characterized in that, The system comprises a data collection module, a ligament graft maturity evaluation module and a tendon-bone interface healing evaluation module. The data collection module is used to collect clinical information and MRI image data of a patient. The ligament graft maturity evaluation module is used to automatically measure and calculate the maturity of an anterior cruciate ligament graft. The tendon-bone interface healing evaluation module is used to automatically measure the changes of bone tunnel diameter and edema around the tunnel. The data collection module is connected with the ligament graft maturity evaluation module and the tendon-bone interface healing evaluation module.

2. The artificial intelligence-based intelligent evaluation system after anterior cruciate ligament reconstruction according to claim 1, characterized in that, The data collection module comprises: a clinical information collection unit for calling basic information and surgical information of a patient; an MRI data collection unit for calling follow-up knee MRI data of a patient.

3. The artificial intelligence-based intelligent evaluation system for anterior cruciate ligament reconstruction after surgery according to claim 2, characterized in that, The ligament graft maturity evaluation module comprises: a ligament graft and background segmentation unit for automatically identifying the anterior cruciate ligament graft and background in the knee MRI data by an artificial intelligence method; a signal measurement unit for measuring the signal intensity of the background and the anterior cruciate ligament graft, respectively; a signal-to-noise ratio calculation unit for calculating the signal-to-noise ratio of the anterior cruciate ligament graft at different positions and evaluating the maturity of the anterior cruciate ligament graft based on the signal-to-noise ratio.

4. The artificial intelligence-based intelligent evaluation system after anterior cruciate ligament reconstruction according to claim 3, characterized in that, The process of measuring the background signal by the signal measurement unit comprises: selecting a knee MRI T2 fat-suppression sequence sagittal position; automatically selecting a circle with a first preset distance in diameter in the segmented background by an artificial intelligence and measuring the signal intensity thereof as the background signal; The method for measuring the signal intensity of the anterior cruciate ligament graft comprises: segmenting the anterior cruciate ligament graft based on a convolutional neural network to obtain a distal end, a middle part and a proximal end of the anterior cruciate ligament graft, and automatically selecting a circle with a first preset distance in diameter in the distal end, the middle part and the proximal end of the anterior cruciate ligament graft and measuring the signal intensity thereof.

5. The artificial intelligence-based intelligent evaluation system for post-anterior cruciate ligament reconstruction according to claim 4, characterized in that, The process of calculating the signal-to-noise ratio of the anterior cruciate ligament graft at different positions by the signal-to-noise ratio calculation unit comprises: subtracting the contrast signal from the signal of any position of the distal end, the middle part or the proximal end of the anterior cruciate ligament graft and dividing by the background signal to obtain the signal-to-noise ratio; wherein a circle with a first preset distance in diameter is automatically selected in a preset position and the signal intensity thereof is measured, and the measurement result is taken as the contrast signal, and the preset position includes quadriceps tendon tissue and posterior cruciate ligament tissue.

6. The artificial intelligence-based intelligent evaluation system after anterior cruciate ligament reconstruction according to claim 1, characterized in that, The tendon-bone interface healing evaluation module comprises: a bone tunnel change detection unit for detecting bone tunnel changes; an edema change detection unit for detecting edema changes around the tunnel.

7. The artificial intelligence-based intelligent evaluation system after anterior cruciate ligament reconstruction according to claim 6, characterized in that, The process of detecting the bone tunnel changes by the bone tunnel change detection unit comprises: selecting a knee MRI T2 fat-suppression sequence axial position and a T2 fat-suppression sequence sagittal position; determining the entry level of the ligament graft at the distal end of the femur and the proximal end of the tibia according to the T2 fat-suppression sequence sagittal position, and positioning the axial position of the femoral tunnel and the tibial tunnel; measuring the tunnel diameter on the T2 axial position layer of the femoral tunnel and the tibial tunnel based on a convolutional neural network. The in-vivo tunnel diameter is called by the data acquisition module, the ratio of the measured tunnel diameter and the in-vivo tunnel diameter is determined to determine the change of the bone tunnel.

8. The artificial intelligence-based intelligent evaluation system after anterior cruciate ligament reconstruction according to claim 7, characterized in that, The process of detecting the change of edema around the tunnel by the tunnel edema change detection unit includes: In the T2 fat suppression sequence of the same plane of the femoral tunnel and the tibial tunnel, the center of the femoral tunnel and the tibial tunnel is respectively taken as the center to draw a circle with a diameter of the second preset distance, the signal intensity is measured and taken as the signal intensity around the bone tunnel; A circle with a diameter of the second preset distance is drawn in the bone edema-free area in the middle of the femur, the marrow signal intensity is measured, and the marrow signal intensity is taken as a reference value; The change of edema around the bone tunnel is determined by the ratio of the signal around the bone tunnel and the reference value. 9.The artificial intelligence-based intelligent anterior cruciate ligament reconstruction postoperative evaluation system according to claim 1, wherein, The system further includes a rehabilitation suggestion module, which includes: A rehabilitation database: used for literature retrieval, research on the correlation between the maturity of the ligament graft after the anterior cruciate ligament reconstruction and the rehabilitation training, rehabilitation methods and actions, and rehabilitation guidelines; An early warning unit: used for providing rehabilitation strategies for patients by the maturity of the ligament graft and the evaluation results of the tendon-bone interface healing, combined with the rehabilitation database.

10. The artificial intelligence-based intelligent evaluation system after anterior cruciate ligament reconstruction according to claim 9, characterized in that, The rehabilitation database can also optimize and update the database content in combination with the rehabilitation results and evaluation results of the patients.

Images