Tibia diaphysis fracture bone healing state evaluation model training method, evaluation method and system

Abstract

This invention provides a training method, evaluation method, and system for assessing the bone healing status of tibial shaft fractures. The method includes: constructing a training sample set, where each sample contains a tibial shaft X-ray image, and adding labels to each sample; constructing an evaluation model, including a detection module and a scoring module, wherein the scoring module no longer uses discrete classification but directly outputs continuous score values ​​through a fully connected layer; the evaluation model takes the tibial shaft X-ray image as input, the detection module detects the predicted fracture area, and the scoring module scores the bone healing status of the predicted fracture area to obtain a predicted score value, which is then used as the output of the evaluation model; the detection module is trained using the training sample set, and after training, the parameters of the detection module are fixed, and then the scoring module is trained again until the preset performance is achieved, thus obtaining the tibial shaft fracture bone healing status assessment model. The evaluation model provided by this invention is highly efficient, highly accurate, and supports automated continuous scoring.

Description

Technical Field

[0001] This invention relates to the field of deep learning technology, and in particular to a training method, evaluation method and system for assessing the bone healing status of tibial shaft fractures. Background Technology

[0002] Tibial shaft fractures are among the most common long bone fractures, and accurate assessment of their healing progress is crucial for developing appropriate rehabilitation plans. For example, determining whether a patient can begin weight-bearing exercises and when to remove internal fixation devices such as plates depends on the assessment of fracture healing status. Currently, physicians primarily use the mRUST (modified Radiographic Union Score for Tibia Fractures) system to assess fracture healing via X-ray imaging. This scoring system requires physicians to independently score multiple cortical bone regions on anteroposterior and lateral X-ray images, which is time-consuming, labor-intensive, and highly subjective. It is particularly inefficient and prone to subjective errors when multiple physicians are needed to collaboratively score the fracture.

[0003] Currently, there is no method for automatic mRUST scoring, but there are some similar studies, such as fracture localization and classification. Aertong et al. used the YOLO model to detect forearm fractures in children, achieving an average accuracy close to that of professional radiologists and significantly better than that of general practitioners. One study indicated that YOLO's average accuracy in fracture detection is superior to other state-of-the-art algorithms. However, YOLO is primarily designed for rapid target detection and performs poorly on complex classification tasks, requiring combination with other networks. Therefore, directly using a one-stage model for fracture localization and mRUST scoring may fail to accurately identify and assess these subtle and gradually changing features, resulting in inaccurate scoring results. In contrast, SwinTransformer performs better in crack feature extraction. Its self-attention mechanism and hierarchical design can capture features of different ranges at different levels, while convolutional networks, by increasing depth and continuous convolutional operations to expand the receptive field, may introduce feature loss. Liu et al. combined YOLO and SwinTransformer for AO / OTA classification of femoral fractures, showing that this method is more accurate than using YOLO alone. However, bone healing requires observation of changes such as callus formation, gradual connection of bone bridges, and blurring of fracture lines. Whether this algorithm is applicable to bone healing research still needs to be verified.

[0004] Bone healing is a dynamic physiological process that involves multiple stages from fracture to complete healing, including the inflammatory phase, the repair phase, and the remodeling phase, involving a series of continuous physiological processes such as cell proliferation, matrix formation, and mineralization. However, traditional classification methods often use the cross-entropy loss function, which ignores the information about the labeling order of different stages of bone healing, thus making the mRUST score unable to accurately reflect the healing status in clinical practice. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a training method, evaluation method and system for assessing the bone healing status of tibial shaft fractures, in order to eliminate or improve one or more defects existing in the prior art.

[0006] On the one hand, the present invention provides a training method for a model for assessing bone healing status in tibial shaft fractures based on mRUST, the method comprising the following steps:

[0007] A training sample set is constructed, which contains multiple samples, each of which contains one X-ray image of the tibial shaft; the actual fracture area and the actual score value are labeled for each X-ray image of the tibial shaft.

[0008] An evaluation model is constructed, which includes a detection module and a scoring module. The scoring module removes the Softmax function and directly outputs the results from the fully connected layer. The evaluation model takes the tibial shaft X-ray image as input. The detection module detects the tibial shaft X-ray image to obtain the predicted fracture area. The scoring module scores the bone healing status of the predicted fracture area to obtain a predicted score value, and the predicted score value is used as the output of the evaluation model.

[0009] The detection module is trained using the training sample set. After training, the parameters of the detection module are fixed, and then the scoring module is trained until the preset performance is achieved, thus obtaining the tibial shaft fracture bone healing status assessment model.

[0010] In some embodiments of the present invention, the tibial shaft X-ray images are preprocessed before constructing the training sample set, including:

[0011] The X-ray image of the tibial shaft is normalized, and the pixel values ​​are scaled to the range of 0 to 1.

[0012] Adaptive histogram equalization is used to increase the contrast of the tibial shaft X-ray images;

[0013] Data augmentation techniques were used to randomly transform the X-ray images of the tibial shaft, generating multiple different versions of the images.

[0014] In some embodiments of the present invention, an evaluation model is constructed, the evaluation model including a detection module and a scoring module, and further comprising:

[0015] The detection module adopts the YOLOv9 model and uses CSPDarkNet as the backbone network. Through programmable gradient information technology, it adaptively adjusts the gradient propagation path, direction and magnitude according to actual needs when calculating gradients, so as to optimize gradient transmission and improve detection accuracy.

[0016] In some embodiments of the present invention, the detection module detects the tibial shaft X-ray image to obtain a predicted fracture region, including:

[0017] The X-ray image of the tibial shaft is input into the detection module;

[0018] The detection module extracts features from the tibial shaft X-ray image through a convolutional network to generate a feature map; anchor boxes are generated on the feature map, and boundary parameters are predicted for each anchor box to obtain the predicted fracture region.

[0019] In some embodiments of the present invention, before the scoring module scores the bone healing status of the predicted fracture area, the method further includes:

[0020] Adjust the pixels of the predicted fracture area. When the size becomes smaller after adjustment, fill the border with black to maintain consistent size.

[0021] In some embodiments of the present invention, training the detection module using the training sample set includes:

[0022] The cross-union ratio (CUI) between the predicted fracture area and the actual fracture area is calculated. The predicted fracture areas are then filtered according to a preset CUI threshold, and the parameters of the detection module are adjusted.

[0023] In some embodiments of the present invention, training the scoring module includes:

[0024] The mean squared error loss between the predicted score and the actual score is constructed, and the scoring module is optimized with the goal of minimizing the mean squared error loss.

[0025] On the other hand, this invention also provides a method for assessing the bone healing status of tibial shaft fractures based on mRUST, the method comprising:

[0026] Obtain X-ray images of the tibial shaft to be evaluated;

[0027] The tibial shaft X-ray image is input into the tibial shaft fracture bone healing status assessment model trained by any of the above-mentioned mRUST-based tibial shaft fracture bone healing status assessment models to generate a tibial shaft fracture bone healing status score.

[0028] On the other hand, this invention also provides a system for assessing the bone healing status of tibial shaft fractures based on mRUST. When executed, the system can realize the method for assessing the bone healing status of tibial shaft fractures based on mRUST as described above. The system includes:

[0029] The data processing module is used to acquire the X-ray image of the tibial shaft to be evaluated and to preprocess the X-ray image of the tibial shaft.

[0030] The status scoring module includes a tibial shaft fracture bone healing status assessment model trained based on any of the above-mentioned mRUST-based tibial shaft fracture bone healing status assessment model training methods, which is used to automatically assess the tibial shaft fracture bone healing status based on the tibial shaft X-ray images.

[0031] The evaluation output module is used to output the score of bone healing status in tibial shaft fractures.

[0032] On the other hand, the present invention also provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of any of the methods mentioned above.

[0033] This invention provides a training method, evaluation method, and system for assessing the bone healing status of tibial shaft fractures, comprising: constructing a training sample set, each sample containing a tibial shaft X-ray image, and adding labels to each sample; constructing an evaluation model, including a detection module and a scoring module, wherein the scoring module no longer uses discrete classification but directly outputs continuous score values ​​through a fully connected layer; the evaluation model takes the tibial shaft X-ray image as input, the detection module detects the predicted fracture area, and the scoring module scores the bone healing status of the predicted fracture area to obtain a predicted score value, which is used as the output of the evaluation model; training the detection module using the training sample set, fixing the parameters of the detection module after training, and then training the scoring module until the preset performance is achieved, thereby obtaining the tibial shaft fracture bone healing status assessment model. The tibial shaft fracture bone healing status assessment model trained based on the training method provided by this invention can accurately detect the fracture area and assess the bone healing status, reduce subjective errors, and improve scoring accuracy. Furthermore, the automated scoring based on this tibial shaft fracture bone healing status assessment model reduces the time and labor costs required for manual scoring and improves scoring efficiency. In addition, by removing the Softmax function of the scoring module and directly outputting the results from the fully connected layer, continuous scoring values ​​can be obtained, thereby reflecting the dynamic process of bone healing in more detail and overcoming the limitation of the discreteness of existing mRUST scoring.

[0034] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the description, or may be learned by practice of the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures specifically pointed out in the description and drawings.

[0035] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0036] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, are not intended to limit the scope of the invention. In the drawings:

[0037] Figure 1 This is a schematic diagram of the steps in a training method for a model to assess the bone healing status of tibial shaft fractures based on mRUST, according to an embodiment of the present invention.

[0038] Figure 2 This is a flowchart illustrating the training method for a tibial shaft fracture bone healing status assessment model based on mRUST in one embodiment of the present invention.

[0039] Figure 3This is a schematic diagram of the structure of the evaluation model in one embodiment of the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0041] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0042] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0043] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0044] In the following description, embodiments of the invention will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0045] To address the shortcomings of existing mRUST scoring techniques for fracture bone healing status, such as low efficiency, high subjectivity, and low automation, and the fact that existing deep learning-based mRUST scoring techniques neglect the label order information of different stages of bone healing, thus leading to the inaccurate clinical reflection of healing status, this invention provides a training method for an mRUST-based model for assessing tibial shaft fracture bone healing status. Figure 1 As shown, the method includes the following steps S101~S103:

[0046] Step S101: Construct a training sample set, which contains multiple samples, each containing a tibial shaft X-ray image; label the actual fracture area and the actual score for each tibial shaft X-ray image.

[0047] Step S102: Construct an evaluation model, which includes a detection module and a scoring module. The scoring module removes the Softmax function and directly outputs the results from the fully connected layer. The evaluation model takes a tibial shaft X-ray image as input. The detection module detects the tibial shaft X-ray image to obtain the predicted fracture area. The scoring module scores the bone healing status of the predicted fracture area to obtain a predicted score value, which is then used as the output of the evaluation model.

[0048] Step S103: Train the detection module using the training sample set. After training, fix the parameters of the detection module and then train the scoring module until the preset performance is achieved, thus obtaining the tibial shaft fracture bone healing status assessment model.

[0049] like Figure 2 The diagram shows the training method for the mRUST-based model for assessing bone healing status in tibial shaft fractures.

[0050] In step S101, a training sample set is constructed.

[0051] In some embodiments, a large number of tibial shaft X-ray images at different healing stages are collected from hospitals. Available tibial shaft X-ray images are selected according to preset criteria to construct a dataset. Multiple experienced orthopedic surgeons annotate the fracture areas and healing status of the tibial shaft X-ray images. When multiple orthopedic surgeons disagree, a more senior orthopedic surgeon makes the final judgment.

[0052] In some embodiments, preprocessing operations are performed on each tibial shaft X-ray image in the dataset, including:

[0053] Normalization is performed on the X-ray images of the tibial shaft, scaling the pixel values ​​to the range of 0 to 1. This avoids problems such as training instability or convergence difficulties caused by different ranges of pixel values ​​during model training.

[0054] Adaptive histogram equalization is used to increase the contrast of tibial shaft X-ray images.

[0055] Data augmentation techniques were used to randomly transform tibial shaft X-ray images, such as rotation, flipping, and scaling, to generate multiple different versions of the images, thereby improving the robustness and generalization ability of the model and reducing the risk of overfitting.

[0056] In some embodiments, the constructed dataset is divided into a training set (training sample set) and a test set according to a preset ratio, such as using 80% of the dataset as the training set and 20% as the test set. The training set is used to train the model, and the test set is used to test the model's performance.

[0057] In step S102, an evaluation model is constructed, which includes a detection module and a scoring module. The detection module is used to detect and locate the fracture area in the preprocessed tibial shaft X-ray image, and the scoring module is used to score the bone healing status of the fracture area.

[0058] In some embodiments, such as Figure 3 As shown, the detection module uses the YOLOv9 model, and the scoring module uses the SwinTransformer V2 model. Specifically, the scoring module needs to remove the Softmax function and directly output the results from the fully connected layer, changing the model from a classification task to a regression task, thereby outputting continuous score values ​​and overcoming the limitation of mRUST score dispersion in existing technologies.

[0059] After preprocessing the tibial shaft X-ray image, including cropping and image enhancement, it is input into the evaluation model. First, the convolutional network of the YOLOv9 model extracts features from the tibial shaft X-ray image, generating a feature map. Anchor boxes are then generated on the feature map, and boundary parameters are predicted for each anchor box to obtain the predicted fracture region. The boundary parameters include the x-coordinate of the center point. y-coordinate of the center point ,width ,high and confidence score The confidence score represents the probability that the anchor box contains the physical object. The intersection-over-union (IoU) ratio between the predicted fracture region and the actual fracture region is calculated. Predicted fracture regions that meet a threshold are selected, and their sizes are adjusted. Figure 3 The predicted fracture area that matches the criteria is represented by ROI (Region of Interest).

[0060] The predicted fracture area detected by the YOLOv9 model is then input into the SwinTransformer V2 model for mRUST scoring. Since the Softmax function is removed, continuous score values ​​are directly output through the fully connected layer to obtain the predicted score value, which is then used as the final output of the evaluation model.

[0061] In some embodiments, the size of the predicted fracture region is adjusted, for example, the predicted fracture region is adjusted from 256x256 pixels to 224x224 pixels, and the size is kept consistent by filling the black border.

[0062] In some embodiments, the detection module uses the YOLOv9 model and CSPDarkNet as the backbone network. Through programmable gradient information technology, the gradient propagation path, direction and magnitude are adaptively adjusted according to actual needs when calculating the gradient, so as to optimize gradient propagation and improve detection accuracy.

[0063] In step S103, the detection module is first trained using X-ray images of the tibial shaft.

[0064] In some embodiments, the cross-union ratio (CUI) between the predicted fracture area and the actual fracture area obtained by the detection module is calculated, the predicted fracture area is filtered according to a preset CUI threshold, and the detection module parameters are adjusted.

[0065] In some embodiments, during the training of the detection module, a stochastic gradient descent optimizer is used with a batch size of 16, an initial learning rate of 0.001, a weight decay of 0.0005, a momentum of 0.937, an image size of 640x640 pixels, and 100 epochs of training. An early stopping strategy is employed to ensure model stability.

[0066] After training is complete, fix the parameters of the detection module, and then train the scoring module.

[0067] In some embodiments, the mean squared error loss between the predicted score value and the actual score value obtained by the scoring module is used to optimize the scoring module with the goal of minimizing the mean squared error loss.

[0068] In some embodiments, the scoring module is trained using the Adam optimizer with a batch size of 32, an initial learning rate of 0.001, for 60 epochs, and an early stopping strategy to prevent overfitting.

[0069] After training, a model for assessing the bone healing status of tibial shaft fractures was finally obtained.

[0070] In some embodiments, the trained tibial shaft fracture bone healing status assessment model is tested using the test set mentioned in step S101 to evaluate the model's performance. Specifically, the YOLOv9 model is evaluated primarily based on macro-average precision, macro-average recall, macro-average F1 score, and average precision under different IoU thresholds; the SwinTransformer V2 model is evaluated primarily based on mean absolute error, mean squared error, root mean square error, and coefficient of determination.

[0071] The present invention also provides a method for assessing the bone healing status of tibial shaft fractures based on mRUST, comprising the following steps S201~S202:

[0072] Step S201: Obtain X-ray images of the tibial shaft to be evaluated.

[0073] Step S202: Input the tibial shaft X-ray image into the tibial shaft fracture bone healing status assessment model trained based on the above-mentioned mRUST-based tibial shaft fracture bone healing status assessment model training method to generate a tibial shaft fracture bone healing status score.

[0074] This invention also provides an mRUST-based system for assessing the bone healing status of tibial shaft fractures. When executed, this system can realize the aforementioned mRUST-based method for assessing the bone healing status of tibial shaft fractures. The system includes:

[0075] The data processing module is used to acquire X-ray images of the tibial shaft to be evaluated and to preprocess the X-ray images of the tibial shaft, such as cropping and image enhancement.

[0076] The status scoring module includes a tibial shaft fracture bone healing status assessment model trained based on the above-mentioned mRUST-based tibial shaft fracture bone healing status assessment model training method, which is used to automatically assess the tibial shaft fracture bone healing status based on tibial shaft X-ray images.

[0077] The evaluation output module is used to output the score of bone healing status in tibial shaft fractures.

[0078] Corresponding to the above method, the present invention also provides an electronic device, which includes a computer device, the computer device including a processor and a memory, the memory storing computer instructions, the processor executing the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the electronic device performs the steps of the method as described above.

[0079] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned edge computing server deployment method. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0080] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave.

[0081] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0082] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0083] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A training method for a model to assess the bone healing status of tibial shaft fractures based on mRUST, characterized in that, The method includes the following steps: A training sample set is constructed, which contains multiple samples, each of which contains one X-ray image of the tibial shaft; the actual fracture area and the actual score value are labeled for each X-ray image of the tibial shaft. An evaluation model is constructed, comprising a detection module and a scoring module. The evaluation model takes the tibial shaft X-ray image as input. The detection module detects the tibial shaft X-ray image to obtain a predicted fracture area. The predicted fracture area is adjusted to fit the input size requirements of the scoring module. The scoring module performs regression analysis on the bone healing status of the adjusted predicted fracture area. The scoring module removes the Softmax function and directly outputs continuous predicted score values ​​from a fully connected layer to continuously quantify the dynamic healing process of the tibial shaft fracture. These predicted score values ​​are then used as the output of the evaluation model. The detection module is trained using the training sample set. After training, the parameters of the detection module are fixed, and then the scoring module is trained until the preset performance is achieved, thus obtaining a tibial shaft fracture bone healing status assessment model. When training the scoring module, the mean squared error loss between the predicted score value and the true score value is constructed, and the scoring module is optimized with the goal of minimizing the mean squared error loss.

2. The training method for the tibial shaft fracture bone healing status assessment model based on mRUST according to claim 1, characterized in that, Before constructing the training sample set, the tibial shaft X-ray images are preprocessed, including: The X-ray image of the tibial shaft is normalized, and the pixel values ​​are scaled to the range of 0 to 1. Adaptive histogram equalization is used to increase the contrast of the tibial shaft X-ray images; Data augmentation techniques were used to randomly transform the X-ray images of the tibial shaft, generating multiple different versions of the images.

3. The training method for the tibial shaft fracture bone healing status assessment model based on mRUST according to claim 1, characterized in that, Construct an evaluation model, which includes a detection module and a scoring module, and also includes: The detection module adopts the YOLOv9 model and uses CSPDarkNet as the backbone network. Through programmable gradient information technology, it adaptively adjusts the gradient propagation path, direction and magnitude according to actual needs when calculating gradients, so as to optimize gradient transmission and improve detection accuracy.

4. The training method for the tibial shaft fracture bone healing status assessment model based on mRUST according to claim 1, characterized in that, The detection module detects the X-ray image of the tibial shaft to obtain the predicted fracture area, including: The X-ray image of the tibial shaft is input into the detection module; The detection module extracts features from the tibial shaft X-ray image through a convolutional network to generate a feature map; anchor boxes are generated on the feature map, and boundary parameters are predicted for each anchor box to obtain the predicted fracture region.

5. The training method for the tibial shaft fracture bone healing status assessment model based on mRUST according to claim 1, characterized in that, Before the scoring module scores the bone healing status of the predicted fracture area, the following steps are also included: Adjust the pixels of the predicted fracture area. When the size becomes smaller after adjustment, fill the border with black to maintain consistent size.

6. The training method for the tibial shaft fracture bone healing status assessment model based on mRUST according to claim 1, characterized in that, When training the detection module using the training sample set, the following steps are included: The cross-union ratio (CUI) between the predicted fracture area and the actual fracture area is calculated. The predicted fracture areas are then filtered according to a preset CUI threshold, and the parameters of the detection module are adjusted.

7. A method for assessing bone healing status in tibial shaft fractures based on mRUST, characterized in that, The method includes: Obtain X-ray images of the tibial shaft to be evaluated; The tibial shaft X-ray image is input into the tibial shaft fracture bone healing status assessment model trained according to the mRUST-based tibial shaft fracture bone healing status assessment model training method according to any one of claims 1 to 6, to generate continuous tibial shaft fracture bone healing status scores.

8. A system for assessing bone healing status in tibial shaft fractures based on mRUST, characterized in that, When the system is executed, it can realize the method for assessing the bone healing status of tibial shaft fractures based on mRUST as described in claim 7. The system includes: The data processing module is used to acquire the X-ray image of the tibial shaft to be evaluated and to preprocess the X-ray image of the tibial shaft. The status scoring module includes a tibial shaft fracture bone healing status assessment model trained based on the mRUST-based tibial shaft fracture bone healing status assessment model training method described in any one of claims 1 to 6, which is used to automatically assess the tibial shaft fracture bone healing status based on the tibial shaft X-ray image. The evaluation output module is used to output continuous scores for the bone healing status of tibial shaft fractures.

9. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method as described in any one of claims 1 to 7.

Images