Automatic generation system and method for tooth extraction diagnosis plan

Abstract

The present application discloses a system and method for automatically generating a tooth extraction diagnosis plan, which relates to the field of tooth extraction diagnosis technology. It collects patient data through a CBCT image data acquisition module, and performs standardization, denoising and contrast enhancement processing through an image preprocessing module to improve image quality. Subsequently, a tooth image segmentation module is used to perform refined segmentation based on boundary enhancement on the preprocessed image to accurately separate the tooth structure. The tooth feature quantification extraction module further extracts multi-dimensional feature parameters from the segmentation results, including key indicators such as root length and curvature. Finally, the tooth extraction intelligent diagnosis module combines the clinical knowledge rule base to automatically give a tooth extraction difficulty assessment and main risk prediction conclusions based on the extracted feature parameters. The structured diagnosis report generation module integrates all relevant information to form a comprehensive diagnostic recommendation, thereby significantly improving the accuracy and efficiency of tooth extraction diagnosis.

Description

Technical Field

[0001] The present application relates to the technical field of tooth extraction diagnosis, and more specifically, to a system and method for automatically generating a tooth extraction diagnosis plan. Background Art

[0002] In the field of oral medicine, especially in the diagnostic process involving complex tooth extraction surgeries, precise image analysis and accurate diagnostic conclusions are crucial to ensuring patient safety and improving treatment outcomes. Traditional tooth extraction diagnosis relies primarily on the doctor's experience and two-dimensional X-rays, a method that has certain limitations, especially when faced with complex tooth positions, root morphology, and relationships with surrounding important structures (such as the alveolar nerve canal or maxillary sinus). With the development of technology, cone beam computed tomography (CBCT) has been widely used in oral medicine due to its high-resolution three-dimensional imaging capabilities. However, how to efficiently and accurately extract useful information from these large amounts of three-dimensional data and make scientific and reasonable diagnostic decisions based on this information remains a challenge.

[0003] Existing technical solutions often face the problem of insufficient feature extraction when processing CBCT images. For example, when analyzing teeth and their surrounding structures, traditional methods may not be able to fully capture subtle anatomical features, such as the specific morphology of the tooth root, the degree of curvature, and important parameters such as bone density, which directly affects the accuracy of the diagnosis. In addition, blurred boundaries are another significant problem. Due to the limited contrast between the teeth and surrounding tissues, coupled with the possible presence of noise interference, it becomes difficult to define the boundaries of the teeth, which in turn affects the subsequent segmentation accuracy and feature quantification process. These problems limit the effectiveness of automated systems in actual clinical applications, resulting in a decrease in the reliability of diagnostic results.

[0004] Therefore, an automatic generation scheme for optimized tooth extraction diagnosis scheme is expected. Summary of the Invention

[0005] Traditional tooth extraction diagnostic methods rely on two-dimensional X-rays, which suffer from image overlap and loss of detailed information. Furthermore, manual analysis of CBCT imaging data is inefficient and highly subjective, making it difficult to accurately quantify tooth characteristic parameters and assess extraction risk. This application proposes a system and method for automatically generating tooth extraction diagnostic plans.

[0006] According to one aspect of the present application, a system for automatically generating a tooth extraction diagnosis plan is provided, comprising: a CBCT image data acquisition module for acquiring CBCT image data of a patient object; an image preprocessing module for performing image preprocessing on the CBCT image data to obtain preprocessed CBCT image data; a tooth image segmentation module for performing refined segmentation based on boundary enhancement on the preprocessed CBCT image data to obtain a tooth image segmentation result map, wherein, in the process of refined segmentation based on boundary enhancement, regional boundaries are highlighted by constructing a directional sensitivity modulation map; a tooth feature quantification extraction module for performing multi-dimensional feature extraction and quantification based on the tooth image segmentation result map to obtain tooth quantitative feature parameters; a tooth extraction intelligent diagnosis module for inputting the tooth quantitative feature parameters into a pre-constructed clinical knowledge rule base to obtain a tooth extraction intelligent diagnosis conclusion; and a structured diagnosis report generation module for integrating the tooth image segmentation result map, the tooth quantitative feature parameters and the tooth extraction intelligent diagnosis conclusion to obtain a structured diagnosis report.

[0007] In the automatic generation system of the tooth extraction diagnosis plan, the image preprocessing module is used to: standardize, denoise and contrast enhance the CBCT image data to obtain the preprocessed CBCT image data.

[0008] In the above-mentioned automatic generation system of tooth extraction diagnosis scheme, the dental image segmentation module includes: a dental image basic segmentation unit, which is used to input the preprocessed CBCT image data into a basic segmentation network to obtain a dental image local visual feature map and a dental image global visual feature map; a dental image multi-level fusion unit, which is used to fuse the dental image local visual feature map and the dental image global visual feature map to obtain a dental image multi-level visual feature map; a dental image boundary significance prediction unit, which is used to input the dental image multi-level visual feature map into a boundary significance prediction network to obtain a boundary significance prediction feature map; a dental image boundary enhancement unit, which is used to fuse the dental image multi-level visual feature map and the boundary significance prediction feature map to obtain a dental image boundary enhancement visual feature map; and a dental image segmentation prediction unit, which is used to input the dental image boundary enhancement visual feature map into a segmentation prediction layer to obtain the dental image segmentation result map.

[0009] In the automatic generation system of the tooth extraction diagnosis scheme, the basic segmentation network includes a hybrid model integrating CNN and Transformer.

[0010] In the automatic generation system of the above-mentioned tooth extraction diagnosis scheme, the tooth image boundary significance prediction unit is used to: set a boundary segmentation threshold; mark all positions in the multi-level visual feature map of the tooth image where the pixel values ​​are greater than the boundary segmentation threshold as 1 and mark the remaining positions as 0 to obtain the boundary significance prediction feature map.

[0011] In the automatic generation system of the above-mentioned tooth extraction diagnosis plan, the tooth image boundary enhancement unit includes: a boundary significance fusion subunit, which is used to fuse the multi-level visual feature map of the tooth image and the boundary significance prediction feature map to obtain a tooth image boundary significance visual feature map; a weighted calculation subunit, which is used to calculate the position-weighted sum between the multi-level visual feature map of the tooth image and the tooth image boundary significance visual feature map to obtain the tooth image boundary enhancement visual feature map.

[0012] In the automatic generation system of the above-mentioned tooth extraction diagnosis scheme, the boundary significance fusion subunit is used to: calculate the position point multiplication between the multi-level visual feature map of the tooth image and the boundary significance prediction feature map to obtain the initial tooth image boundary significance visual feature map; calculate the directional gradient amplitude feature map of the multi-level visual feature map of the tooth image relative to the boundary significance prediction feature map; calculate the partial derivatives of the eigenvalues ​​of each position in the directional gradient amplitude feature map relative to the eigenvalues ​​of each position in the multi-level visual feature map of the tooth image to obtain a first bidirectional partial derivative structure interaction coefficient modulation map; calculate the partial derivatives of the eigenvalues ​​of each position in the directional gradient amplitude feature map relative to the eigenvalues ​​of each position in the boundary significance prediction feature map to obtain a second bidirectional partial derivative structure interaction coefficient modulation map; construct a directional sensitivity modulation map based on the first bidirectional partial derivative structure interaction coefficient modulation map and the second bidirectional partial derivative structure interaction coefficient modulation map; calculate the position point multiplication between the directional sensitivity modulation map and the initial tooth image boundary significance visual feature map to obtain the tooth image boundary significance visual feature map.

[0013] In the automatic generation system of the above-mentioned tooth extraction diagnosis plan, the tooth feature quantitative extraction module includes: a tooth structure morphology measurement unit, which is used to perform structure morphology measurement on the tooth image segmentation result image to obtain root length, diameter, curvature, alveolar bone height and bone density estimation; a tooth spatial relationship calculation unit, which is used to perform spatial relationship calculation on the tooth image segmentation result image to obtain the shortest distance between the root tip and the inferior alveolar nerve canal or the maxillary sinus floor; a tooth lesion feature analysis unit, which is used to perform lesion feature analysis on the tooth image segmentation result image to obtain the volume, position and grayscale features of apical lesions, cysts, etc.

[0014] In the automatic generation system of the above-mentioned tooth extraction diagnosis plan, the tooth extraction intelligent diagnosis module is used to: input the quantitative characteristic parameters of the tooth into a pre-built clinical knowledge rule base to obtain the tooth extraction difficulty assessment result and the main risk prediction result, and the tooth extraction difficulty assessment result and the main risk prediction result constitute the tooth extraction intelligent diagnosis conclusion.

[0015] According to another aspect of the present application, a method for automatically generating a tooth extraction diagnosis plan is also provided, which uses the above-mentioned automatic generation system for tooth extraction diagnosis plans, and the method includes: acquiring CBCT image data of a patient object; performing image preprocessing on the CBCT image data to obtain preprocessed CBCT image data; performing fine segmentation based on boundary enhancement on the preprocessed CBCT image data to obtain a tooth image segmentation result map, wherein, in the process of fine segmentation based on boundary enhancement, regional boundaries are highlighted by constructing a directional sensitivity modulation map; multi-dimensional feature extraction and quantification are performed based on the tooth image segmentation result map to obtain tooth quantitative feature parameters; the tooth quantitative feature parameters are input into a pre-constructed clinical knowledge rule base to obtain a tooth extraction intelligent diagnosis conclusion; the tooth image segmentation result map, the tooth quantitative feature parameters and the tooth extraction intelligent diagnosis conclusion are integrated to obtain a structured diagnosis report.

[0016] Compared with the existing technology, the automatic generation system and method of tooth extraction diagnosis scheme provided by this application collects patient data through the CBCT image data acquisition module, and performs standardization, denoising and contrast enhancement processing through the image preprocessing module to improve image quality. Subsequently, the tooth image segmentation module is used to perform refined segmentation based on boundary enhancement on the preprocessed image to accurately separate the tooth structure. The tooth feature quantitative extraction module further extracts multi-dimensional feature parameters from the segmentation results, including key indicators such as root length and curvature. Finally, the tooth extraction intelligent diagnosis module combines the clinical knowledge rule base to automatically give a tooth extraction difficulty assessment and main risk prediction conclusions based on the extracted feature parameters. The structured diagnosis report generation module integrates all relevant information to form a comprehensive diagnostic recommendation, thereby significantly improving the accuracy and efficiency of tooth extraction diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the examples of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the examples of the present application and constitute a part of the specification. Together with the examples of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 The figure shows a schematic block diagram of a system for automatically generating a tooth extraction diagnosis plan according to an example of the present application.

[0019] Figure 2 The figure shows a schematic block diagram of a tooth image segmentation module in the automatic generation system of a tooth extraction diagnosis plan according to an example of the present application.

[0020] Figure 3 The figure shows a schematic block diagram of a tooth image boundary enhancement unit in the automatic generation system of tooth extraction diagnosis plan according to an example of the present application.

[0021] Figure 4 The figure illustrates a schematic block diagram of a tooth feature quantitative extraction module in an automatic generation system of a tooth extraction diagnosis plan according to an example of the present application.

[0022] Figure 5 The figure shows a schematic flow chart of a method for automatically generating a tooth extraction diagnosis plan according to an example of the present application. DETAILED DESCRIPTION

[0023] Below, examples of the present application will be described in detail with reference to the accompanying drawings. Obviously, the described examples are only part of the examples of the present application, rather than all the examples of the present application, and it should be understood that the present application is not limited to the examples described here.

[0024] Figure 1 FIG2 shows a schematic flow chart of an automatic generation system of a tooth extraction diagnosis plan according to an example of the present application. Figure 1 As shown, the present application provides an automatic generation system 10 for a tooth extraction diagnosis scheme, comprising: a CBCT image data acquisition module 11 for acquiring CBCT image data of a patient object; an image preprocessing module 12 for performing image preprocessing on the CBCT image data to obtain preprocessed CBCT image data; a tooth image segmentation module 13 for performing fine segmentation based on boundary enhancement on the preprocessed CBCT image data to obtain a tooth image segmentation result map, wherein, in the process of fine segmentation based on boundary enhancement, regional boundaries are highlighted by constructing a directional sensitivity modulation map; a tooth feature quantification extraction module 14 for performing multi-dimensional feature extraction and quantification based on the tooth image segmentation result map to obtain tooth quantitative feature parameters; a tooth extraction intelligent diagnosis module 15 for inputting the tooth quantitative feature parameters into a pre-constructed clinical knowledge rule base to obtain a tooth extraction intelligent diagnosis conclusion; and a structured diagnosis report generation module 16 for integrating the tooth image segmentation result map, the tooth quantitative feature parameters and the tooth extraction intelligent diagnosis conclusion to obtain a structured diagnosis report.

[0025] Specifically, the CBCT image data acquisition module 11 acquires CBCT image data from a patient. This process is performed in a professional dental clinic or hospital by experienced radiologists operating a cone-beam computed tomography (CBCT) device. First, the patient follows instructions to assume an appropriate position, typically sitting in a custom chair with their head fixed in a specific position to minimize motion artifacts. CBCT device parameters, such as voltage, current, and exposure time, are then adjusted based on the patient's specific circumstances (e.g., age, body mass index, and examination site), optimizing image quality while minimizing radiation dose.

[0026] In a specific example, when diagnosing a young adult who needs to have his mandibular wisdom teeth removed, the technician will select an appropriate scanning range, focusing on the mandible and its surrounding structures, including the inferior alveolar nerve canal. During this process, the CBCT device will rotate around the patient's head, emitting X-rays and receiving the detector after attenuation by different tissues. These raw data are then transmitted to the computer system. It is worth noting that in order to ensure the integrity and accuracy of the image data, various interference factors in the scanning environment must be strictly controlled, such as electromagnetic interference, temperature changes, etc. This can be achieved through the use of shielding materials and technical means.

[0027] In a preferred example, taking into account the diversity and complexity of practical applications, obtaining CBCT imaging data of patient subjects also includes how to effectively manage and store these huge data sets. Specifically, on the one hand, with the development of digital technology, medical institutions can now use cloud storage solutions to safely store patients' CBCT imaging data, which not only facilitates cross-departmental and even cross-border collaboration and communication, but also facilitates long-term tracking and observation of disease progression. On the other hand, for scenarios with high real-time requirements, such as rapid assessment before emergency surgery, it may be necessary to deploy high-performance computing resources to speed up the image reconstruction process to ensure that doctors can obtain clear and accurate diagnostic basis in the shortest possible time.

[0028] Specifically, in the image preprocessing module 12, the CBCT image data is preprocessed to obtain preprocessed CBCT image data. It should be understood that in the field of oral medicine, especially for tooth extraction diagnosis, it is crucial to obtain high-quality CBCT image data. However, the original CBCT image data often has problems such as noise and low contrast. If these problems are not handled, they will directly affect the subsequent tooth image segmentation, feature extraction and the final intelligent diagnosis results. Therefore, in one example, the image preprocessing module 12 is used to: standardize, denoise and contrast enhance the CBCT image data to obtain the preprocessed CBCT image data.

[0029] Specifically, standardization ensures consistency across CBCT image data collected from different sources and under different conditions. Images captured using different equipment and settings may exhibit variations in brightness and grayscale. These differences, if not eliminated, can lead to inaccurate analysis results. Standardization adjusts the image's grayscale range to conform to a predetermined standard range, such as 0 to 255. This ensures that all images input to the system are on the same scale, facilitating unified processing.

[0030] At the same time, denoising is one of the key steps to improve image quality. The noise in CBCT image data may come from a variety of factors, such as electronic noise of the equipment itself, artifacts caused by patient movement, etc. Removing these noises not only helps to improve the clarity of the image, but also reduces the possibility of misjudgment in the subsequent processing stage. Commonly used denoising methods include Gaussian filtering and median filtering. For example, Gaussian filtering can smooth out some random noise while retaining important details; median filtering is particularly suitable for removing salt and pepper noise. It achieves noise reduction by replacing the central pixel value with the median of the neighboring pixel values.

[0031] Contrast enhancement is crucial for highlighting key structures in images. Good contrast makes it easier for doctors or automated systems to distinguish different tissue types, such as the boundary between teeth and surrounding bone tissue. Common contrast enhancement techniques include histogram equalization and adaptive contrast enhancement. Histogram equalization expands the dynamic range by redistributing the grayscale values ​​of the image, making the details in the image more clearly visible; adaptive contrast enhancement adjusts the contrast based on the characteristics of the local area and is particularly suitable for processing images with large brightness changes.

[0032] In a specific example, after the patient's CBCT image data is imported into the system, it is first converted into a standard format through a standardization algorithm to ensure that all subsequent operations are based on a consistent basis. Next, a denoising algorithm is applied to remove various possible interference factors, such as using a Gaussian filter to smooth the image and reduce unnecessary noise. Subsequently, histogram equalization or adaptive contrast enhancement technology is used to improve the overall contrast of the image, making the teeth and their surrounding structures more distinct, facilitating accurate segmentation and feature extraction by subsequent modules. Through such a series of meticulous preprocessing steps, not only can the quality of the image be significantly improved, but it can also provide more reliable data support for subsequent intelligent diagnosis, thereby improving the accuracy and efficiency of the entire tooth extraction diagnosis plan generation system.

[0033] Specifically, in the dental image segmentation module 13, the pre-processed CBCT image data is subjected to refined segmentation based on boundary enhancement to obtain a dental image segmentation result map. It should be understood that traditional single models are often difficult to meet the requirements of high-precision segmentation at the same time. For example, relying solely on CNN may not be able to fully consider global information, while using Transformer alone may lead to excessive consumption of computing resources. In contrast, the hybrid model that integrates CNN and Transformer can not only efficiently extract local details but also take into account global structural information, and is very suitable for segmentation tasks of objects with complex geometric characteristics such as teeth. In addition, through strategies such as multi-level feature fusion and boundary saliency enhancement, the quality of segmentation results can be significantly improved, the misjudgment rate can be reduced, and more reliable support can be provided for subsequent diagnosis.

[0034] In one example, if Figure 2 As shown, the dental image segmentation module 13 includes: a dental image basic segmentation unit 131, which is used to input the preprocessed CBCT image data into a basic segmentation network to obtain a dental image local visual feature map and a dental image global visual feature map; a dental image multi-level fusion unit 132, which is used to fuse the dental image local visual feature map and the dental image global visual feature map to obtain a dental image multi-level visual feature map; a dental image boundary significance prediction unit 133, which is used to input the dental image multi-level visual feature map into a boundary significance prediction network to obtain a boundary significance prediction feature map; a dental image boundary enhancement unit 134, which is used to fuse the dental image multi-level visual feature map and the boundary significance prediction feature map to obtain a dental image boundary enhancement visual feature map; and a dental image segmentation prediction unit 135, which is used to input the dental image boundary enhancement visual feature map into a segmentation prediction layer to obtain the dental image segmentation result map.

[0035] Specifically, first, the preprocessed CBCT image data is input into a basic segmentation network. In one example, the basic segmentation network includes a hybrid model integrating CNN and Transformer. Specifically, CNN is good at capturing local detail information in the image, which is crucial for accurately identifying the specific shape, size and position of teeth. On the other hand, Transformer is known for its powerful global context understanding ability. It can effectively model long-range dependencies in the entire image, which helps to more accurately define the boundaries between teeth and other structures. In a specific example, the data representation of the input image is initialized through a multi-layer perceptron (MLP) embedding layer, and then the feature representation is gradually refined through a series of alternating stacked CNN blocks and Transformer encoder layers.

[0036] Those skilled in the art will understand that a CNN block typically consists of a series of convolutional layers, activation functions, and pooling layers. The convolutional layer, the core component of a CNN, utilizes a set of learnable filters (or kernels) that slide over the input data to detect different local features, such as edges and textures. These filters are trained using a backpropagation algorithm, gradually optimizing their parameters to more accurately recognize specific patterns. Activation functions, such as ReLU (Rectified Linear Unit), are applied after each convolutional layer to introduce nonlinearity, which is crucial for modeling complex input-output relationships. Furthermore, pooling layers, especially max pooling layers, are often used to reduce the spatial size of feature maps, thereby reducing computational complexity and mitigating the risk of overfitting. Through this design, a CNN block can efficiently compress information without losing critical details, providing a concise yet informative feature representation for subsequent processing. The Transformer encoder layer, on the other hand, employs a completely different mechanism to process sequential data, particularly for tasks that require understanding global context. A core concept of the Transformer architecture is the self-attention mechanism, which allows the model to consider information about all other elements in a sequence for each element. This means that for given dental CBCT image data, Transformer can dynamically adjust the weights between different regions to ensure that even structures far away from the current focus receive appropriate attention. This feature enables Transformer to perform well when processing objects with complex geometric shapes, because it can not only capture local details but also understand the macroscopic structure of the entire image. Specifically, the Transformer encoder layer consists of two parts: a multi-head self-attention mechanism and a feedforward neural network. The former is responsible for generating query, key, and value matrices, and determining the importance of each position by calculating the similarity score between them; the latter further processes these weighted feature representations to enhance the learning ability of the model.

[0037] Next, after obtaining the local and global visual feature maps, they need to be fused together to form a multi-level visual feature map. An attention-based approach is employed here, enabling the model to dynamically adjust its focus at different scales, thereby better capturing subtle structural changes in the teeth. In a specific example, the similarity matrix between two feature maps is calculated, converted into a weight matrix using the softmax function, and then weighted summed to obtain the final multi-level visual feature map. This approach allows the model to automatically focus on the most informative parts, improving segmentation accuracy.

[0038] Next, to further clarify the tooth boundaries, a boundary saliency prediction network is used to generate a boundary saliency prediction feature map. In one example, the tooth image boundary saliency prediction unit 133 is configured to set a boundary segmentation threshold, marking all pixel locations in the tooth image multi-level visual feature map with values ​​greater than the boundary segmentation threshold as 1, and marking all other locations as 0, to generate the boundary saliency prediction feature map. It should be understood that the boundary segmentation threshold is determined based on existing medical knowledge and empirical data. For example, in common CBCT images, teeth (especially enamel) may have relatively high grayscale values, ranging from approximately 1500 to 3000 Hounsfield units (HU). Bone tissue, on the other hand, has a relatively low grayscale value, but still significantly higher than soft tissue, roughly ranging from 500 to 1200 HU. Based on this, the present application selects a fixed threshold between these two thresholds for segmentation. In one specific example, 1200 HU is used as the boundary segmentation threshold. This means that in subsequent processing, for each pixel in the multi-level visual feature map, if its grayscale value is greater than 1200HU, the pixel is considered to belong to the tooth area and is marked as 1; conversely, if the grayscale value is less than or equal to 1200HU, it is considered to be the background (i.e., a non-tooth area) and is marked as 0. Of course, the above is only an example and this example is not specific.

[0039] However, directly applying a fixed threshold can lead to edge blur or discontinuity. Therefore, in a preferred example, an adaptive threshold adjustment strategy, such as the Otsu algorithm, is introduced. This algorithm automatically selects the optimal threshold based on the image histogram, ensuring boundary clarity and consistency. Specifically, the grayscale values ​​of all pixels to be processed are collected and a histogram is constructed. This histogram reflects the distribution of pixels at different grayscale levels. Next, the Otsu algorithm is applied to calculate the optimal threshold. The core idea of ​​this algorithm is to find a threshold that maximizes the inter-class variance between the foreground and background. In other words, the goal is to find a grayscale level that separates the two components (foreground and background) as closely as possible while maintaining internal consistency. Specifically, the Otsu algorithm iterates through all possible thresholds. For each threshold, it calculates the average grayscale value of the foreground and background and, based on this, calculates the inter-class variance. Ultimately, the threshold that maximizes the inter-class variance is selected as the boundary segmentation threshold.

[0040] Finally, the tooth image boundary is enhanced by fusing the multi-level visual feature map and the boundary saliency prediction feature map. Figure 3 As shown, the tooth image boundary enhancement unit 134 includes: a boundary significance fusion subunit 1341, used to fuse the tooth image multi-level visual feature map and the boundary significance prediction feature map to obtain a tooth image boundary significance visual feature map; a weighted calculation subunit 1342, used to calculate the position-weighted sum between the tooth image multi-level visual feature map and the tooth image boundary significance visual feature map to obtain the tooth image boundary enhancement visual feature map.

[0041] Here, considering that when fusing the multi-level visual feature map of the dental image and the boundary saliency prediction feature map, the initial dental image boundary saliency visual feature map is first obtained by multiplying the multi-level visual feature map of the dental image and the boundary saliency prediction feature map by position points, so as to perform boundary saliency prediction of the multi-level visual features of the dental image data. However, it is obvious that in the actual visual feature representation of the multi-level visual feature map of the dental image, the visual feature salient area may have obvious extreme value distribution characteristics of the gradient amplitude when cross-boundary segmentation occurs, thereby affecting the adaptive sensitivity of the regional salient edge on the overall feature map.

[0042] Based on this, the multi-level visual feature map of the tooth image is represented as , and the boundary saliency prediction feature map is expressed as First, the directional gradient amplitude feature map of the multi-level visual feature map of the tooth image relative to the boundary saliency prediction feature map is calculated, which is expressed as:

[0043] ;

[0044] in, Indicates subtraction by position, Indicates point multiplication by position, Represents the oriented gradient magnitude feature map.

[0045] That is, the multi-level visual feature map of the tooth image is obtained by the above formula Saliency prediction feature map relative to the boundary The directional gradient amplitude is defined, so that the gradient field direction distribution is used as the core constraint condition of significance. Then, for the extreme value distribution, its partial derivatives are calculated respectively. Specifically, the partial derivatives of the eigenvalues ​​at each position in the directional gradient amplitude feature map relative to the eigenvalues ​​at each position in the multi-level visual feature map of the tooth image are calculated to obtain a first bidirectional partial derivative structure interaction coefficient modulation map, and the partial derivatives of the eigenvalues ​​at each position in the directional gradient amplitude feature map relative to the eigenvalues ​​at each position in the boundary saliency prediction feature map are calculated to obtain a second bidirectional partial derivative structure interaction coefficient modulation map, which is expressed as:

[0046] ;

[0047] ;

[0048] in represents the calculation of the inverse of the eigenvalue of each position in the multi-level visual feature map of the tooth image, It represents calculating the square of the reciprocal of the eigenvalue of each position in the multi-level visual feature map of the tooth image.

[0049] That is, by constructing a bidirectional partial derivative structure interaction, the gradient information of the gradient field is associated with the cross-boundary segmentation distribution of multi-level visual features in a differentiable extreme value sampling manner.

[0050] Then, based on the first bidirectional partial guide structure interaction coefficient modulation map and the second bidirectional partial guide structure interaction coefficient modulation map, a directional sensitivity modulation map is constructed, which is expressed as:

[0051] ;

[0052] in, represents a learnable hyperparameter, It means adding by position. Represents the directional sensitivity modulation map.

[0053] By constructing a directional sensitivity modulation map, the region boundary is highlighted, that is, adaptive normalization is performed for directional sensitivity. Here you can learn hyperparameters , through its introduction, the boundary direction information is made consistent with the gradient information.

[0054] Finally, the tooth image boundary salient visual feature map is obtained by calculating the positional multiplication between the directional sensitivity modulation map and the initial tooth image boundary salient visual feature map. In this way, when the multi-level tooth image visual feature map is further fused with the tooth image boundary salient visual feature map, adaptive salient sensitivity consistency between salient distinguishable features can be achieved based on gradient directional sensitivity, thereby improving the expressive effect of the tooth image boundary enhancement visual feature map.

[0055] On this basis, the present application also uses a segmentation prediction layer to perform the final segmentation operation on the enhanced feature map. The segmentation prediction layer is usually composed of several convolutional layers, and its function is to classify the tooth image boundary enhanced visual feature map pixel by pixel, determine whether each pixel belongs to the background or the target area, and obtain the tooth image segmentation result map. In one example, the tooth image boundary enhanced visual feature map is input into the segmentation prediction layer to obtain the tooth image segmentation result map, including: first, using a series of 1x1 convolution kernels to reduce the number of channels of the feature map, and adjusting the feature map through this convolution operation to prepare for classification. The 1x1 convolution here plays the role of dimensionality reduction and feature reorganization, which helps to simplify the subsequent calculation process. Then, the Softmax function is used to classify each pixel. Softmax assigns a probability value of belonging to the target category to each pixel, so that the probability distribution of each pixel point belonging to the tooth can be obtained. The category with the highest probability is selected as the label of the pixel point, thereby completing the segmentation. After completing the above processing, the segmentation prediction layer directly outputs the final tooth image segmentation result map, which clearly defines the boundary between the teeth and the background area, providing an accurate basis for the subsequent quantitative extraction of tooth features.

[0056] Specifically, in the tooth feature quantification extraction module 14, multi-dimensional feature extraction and quantification are performed based on the tooth image segmentation result map to obtain tooth quantitative feature parameters. It should be understood that the specific situation of each patient is different, so personalized treatment plans are particularly important. Multi-dimensional feature extraction and quantification can reveal differences between individuals, thereby supporting more accurate diagnosis and treatment decisions. For example, for patients with specific diseases (such as osteoporosis), their bone density may be lower than the normal range, which requires adjustment of the surgical plan to ensure safety and effectiveness. By conducting a comprehensive analysis of the teeth and their surrounding structures, doctors can tailor the most appropriate treatment strategy according to the specific situation of each patient, improving treatment effectiveness while reducing risks.

[0057] In one example, if Figure 4As shown, the tooth feature quantitative extraction module 14 includes: a tooth structure morphology measurement unit 141, which is used to perform structure morphology measurement on the tooth image segmentation result image to obtain root length, diameter, curvature, alveolar bone height and bone density estimation; a tooth space relationship calculation unit 142, which is used to perform space relationship calculation on the tooth image segmentation result image to obtain the shortest distance between the root tip and the inferior alveolar nerve canal or the maxillary sinus floor; a tooth lesion feature analysis unit 143, which is used to perform lesion feature analysis on the tooth image segmentation result image to obtain the volume, position and grayscale features of apical lesions, cysts, etc.

[0058] In a specific example, when measuring tooth structure morphology, geometric calculation methods can be used to obtain parameters such as root length, diameter, and curvature. For example, root length can be measured by detecting two endpoints on the tooth contour (i.e., the crown apex and the root end) and then calculating the Euclidean distance between these two points. This method is simple and intuitive and suitable for most cases. To improve accuracy, edge detection techniques such as the Canny detector can be combined to first accurately locate the tooth contour and then calculate the length based on the characteristic points on the contour. Similarly, root diameter can be estimated by measuring the maximum and minimum diameters at a specific cross-section, while curvature can be approximated by fitting a curve to represent the curved shape of the tooth and calculating the curvature of this curve as a measurement standard. These geometric calculations are not only easy to implement but also can quickly provide preliminary measurement results, laying the foundation for further refinement.

[0059] Secondly, calculating the spatial relationship of teeth, particularly determining the distance between the root apex and important anatomical structures (such as the inferior alveolar nerve canal or the maxillary sinus floor), requires a precise spatial positioning system. Three-dimensional reconstruction techniques can be used to convert the two-dimensional segmentation image into a three-dimensional model. Based on this three-dimensional model, a nearest neighbor search algorithm can be used to find the shortest Euclidean distance between these points of interest. For example, the specific coordinates of the inferior alveolar nerve canal or the maxillary sinus floor can be defined first, and then the straight-line distance between these points and the root apex can be calculated using a distance formula in three-dimensional space. Furthermore, to improve computational efficiency, a small lookup table or hash table can be pre-trained specifically for quickly finding nearest neighbor points. This approach ensures computational speed while maintaining a high level of accuracy. This geometric and mathematical approach can efficiently solve practical problems without relying on complex models, demonstrating its practicality and flexibility.

[0060] Finally, the analysis of dental lesion characteristics involves the identification and quantification of pathological features such as apical lesions and cysts. This step typically requires more detailed analysis methods. One feasible solution is to use traditional image processing techniques, such as region growing or watershed algorithms, to segment the lesion area. Next, the volume of the lesion area is estimated by counting the number of pixels in the lesion area. Simultaneously, color space conversion techniques are used to convert the original RGB image to a color space more suitable for analyzing lesion characteristics (such as HSV), thereby extracting the color and grayscale features of the lesion area. Furthermore, threshold segmentation techniques can be introduced to distinguish the lesion area from other normal tissue based on its unique grayscale range. Although relatively simple, this method performs well in processing specific types of lesions, especially in scenarios where the data volume is small or large-scale automation is not required.

[0061] Furthermore, considering that traditional manual measurement methods are difficult to cope with complex tooth morphologies and pathological conditions, and a single automated tool may not perform well on certain specific tasks. In contrast, this application also proposes a multi-level, multi-modal fusion deep learning model that can fully utilize the advantages of modern computer vision technology to achieve efficient and accurate extraction of tooth features. At the same time, by integrating a variety of advanced technical means, such as three-dimensional reconstruction, nearest neighbor search, and transfer learning, it not only enhances the functional diversity of the system, but also improves the stability and reliability of the overall performance.

[0062] In a preferred example, a method combining convolutional neural networks (CNNs) and traditional image processing techniques can be employed to measure tooth structure morphology. For example, the U-Net architecture, widely used in medical image analysis for its superior semantic segmentation capabilities, can be used as the underlying model. U-Net effectively captures tooth boundary information and generates high-quality segmentation images. Based on this segmentation image, traditional edge detection algorithms, such as the Canny detector, are further utilized to precisely locate tooth contours. Geometric calculations are then applied to measure parameters such as root length, diameter, and curvature. Furthermore, 3D reconstruction techniques can be incorporated to construct a 3D tooth model from the 2D segmentation image, enabling more accurate estimation of metrics such as alveolar bone height and bone density. This approach leverages the powerful feature extraction capabilities of deep learning models while combining the efficiency of traditional algorithms for specific tasks, ensuring accurate measurement results.

[0063] Secondly, calculating the spatial relationships of teeth, particularly measuring the distance between the root tip and key anatomical structures (such as the inferior alveolar canal or the maxillary sinus floor), requires a precise spatial positioning system. A model based on the Transformer architecture can be considered to enhance global context understanding, as the Transformer excels at handling long-range dependencies, which is particularly important for calculating spatial distances across regions. Specifically, a Transformer encoder layer can be added to the aforementioned U-Net to capture global information from the entire image. Next, by defining specific points of interest (POIs), such as the inferior alveolar canal or the maxillary sinus floor, a nearest neighbor search algorithm is used to find the shortest Euclidean distance between these POIs and the root tip. To improve computational efficiency, a small pre-trained neural network can be used specifically for fast nearest neighbor search, ensuring computational speed while maintaining a high level of accuracy.

[0064] Finally, in terms of the analysis of dental lesion characteristics, it involves the identification and quantification of pathological features such as apical lesions and cysts. In the specific example of this application, the Mask R-CNN framework is used, which is a powerful tool designed specifically for instance segmentation tasks. Mask R-CNN can not only accurately locate the lesion area, but also generate an accurate mask for it, which is convenient for subsequent quantitative analysis. On this basis, the volume of the lesion area can be estimated by counting the number of pixels in the lesion area; at the same time, the color space conversion technology is used to convert the original RGB image into a color space (such as HSV) that is more suitable for analyzing the characteristics of the lesion, and then the color and grayscale features of the lesion area are extracted. In addition, a transfer learning strategy can be introduced, that is, the network is initialized using the pre-trained model weights, and then fine-tuned for the specific lesion type. This method not only speeds up the training process, but also improves the adaptability and generalization ability of the model to different cases.

[0065] Specifically, in the tooth extraction intelligent diagnosis module 15, the quantitative characteristic parameters of the teeth are input into a pre-built clinical knowledge rule base to obtain a tooth extraction intelligent diagnosis conclusion. It should be understood that traditional diagnostic methods often rely on the doctor's personal experience and technical level and are easily affected by subjective factors. The method based on the clinical knowledge rule base proposed in this application attempts to overcome these problems by integrating multi-source data and expert wisdom to establish an objective and systematic decision support framework. On the one hand, this approach helps to standardize the diagnosis and treatment process and reduce the probability of human error; on the other hand, with the help of advanced data analysis technology and automation tools, work efficiency can be significantly improved, so that even inexperienced medical staff can make reasonable treatment choices. In addition, considering the dynamic and diverse nature of the medical environment, the design of the rule base should also have good scalability and flexibility so as to absorb new research results and practical experience at any time.

[0066] In one example, the tooth extraction intelligent diagnosis module 15 is used to: input the tooth quantitative characteristic parameters into a pre-built clinical knowledge rule base to obtain a tooth extraction difficulty assessment result and a major risk prediction result, and the tooth extraction difficulty assessment result and the major risk prediction result constitute the tooth extraction intelligent diagnosis conclusion.

[0067] In a specific example, building a clinical knowledge rule base involves: first, collecting data from multiple sources, including but not limited to electronic health records (EHRs), clinical trial results, professional literature, and feedback from actual operations. For example, in the initial construction stage, data mining technology can be used to extract case data related to tooth extraction from the existing EHR database. This data covers information on multiple dimensions such as the patient's age, gender, dental condition, and previous medical history. Next, experts in the field are invited to participate in the design of the rule base. Based on their own clinical experience and the latest research results, they define a series of rules to guide diagnostic decisions under different feature combinations. For example, for specific types of dental lesions or anatomical abnormalities, corresponding treatment recommendations and expected risk levels are specified. In addition, machine learning algorithms can be used to optimize and adjust the above rules to ensure that the rule base can adapt to the ever-changing needs of clinical practice.

[0068] In practical applications, after the quantitative feature parameters of teeth are extracted, they will be passed as input to this knowledge rule base. The key here is how to effectively map these parameters to specific diagnostic conclusions. To this end, a method based on fuzzy logic reasoning can be used. This method allows for handling uncertainty and imprecision and is very suitable for medical decision support systems. For example, assuming that a patient has a more severe root curvature and a lower bone density estimate, then according to the relevant rules in the rule base, the system may infer a higher tooth extraction difficulty score and predict the main risks that may arise, such as postoperative infection or bleeding. At the same time, in order to improve the efficiency of reasoning, an index mechanism or hash table can be introduced in the rule base to quickly locate the best rule set applicable to the current feature combination.

[0069] Specifically, in the structured diagnostic report generation module 16, the tooth image segmentation result map, the tooth quantitative characteristic parameters and the tooth extraction intelligent diagnosis conclusion are integrated to obtain a structured diagnostic report. It should be understood that traditional unstructured reports often lack unified standards, which can easily lead to information omissions or interpretation deviations. The present application solves this problem by integrating the tooth image segmentation result map, the tooth quantitative characteristic parameters and the tooth extraction intelligent diagnosis conclusion to obtain a structured diagnostic report. On the one hand, by adopting a standardized template, it can ensure that the format of the report generated each time is consistent, making it convenient for doctors to quickly obtain the required information; on the other hand, with the help of modern information technology, such as image processing technology and the application of database management systems, not only can work efficiency be improved, but also the authenticity and integrity of the data can be guaranteed.

[0070] In a specific example, a structured diagnostic report consists of the following main sections: 1. Patient Basic Information: This includes basic information such as the patient's name, age, gender, and medical record number. This section primarily identifies the report's subject and ensures accuracy. 2. Dental Image Segmentation Results: This section embeds dental image segmentation results images into the report. These images should clearly indicate the specific location of the teeth and the important anatomical structures surrounding them. To facilitate viewing, interactive features such as zooming in, zooming out, and rotating can be added, allowing the physician to view the teeth and surrounding area from multiple angles. Furthermore, areas of particular concern, such as those with potential lesions, can be highlighted to help the physician quickly locate key points. 3. Dental Quantitative Feature Parameters: This section lists precisely calculated dental quantitative feature parameters, such as root length, diameter, curvature, alveolar bone height, and bone density estimates. To enhance comprehensibility, in addition to providing specific values, reference values ​​within the normal range should be included for comparison, making abnormalities readily apparent. 4. Intelligent Tooth Extraction Diagnosis Conclusion: The intelligent tooth extraction diagnosis conclusion, derived from the clinical knowledge rule base, should be detailed, including but not limited to the extraction difficulty assessment and key risk prediction results. 5. Additional Notes and Suggestions: Finally, space is reserved for the physician to add personal insights or other considerations. This not only facilitates the development of personalized treatment plans but also provides a basis for subsequent review.

[0071] In summary, the automatic generation system of tooth extraction diagnosis scheme provided by this application collects patient data through the CBCT image data acquisition module, and performs standardization, denoising and contrast enhancement processing through the image preprocessing module to improve image quality. Subsequently, the tooth image segmentation module is used to perform refined segmentation based on boundary enhancement on the preprocessed image to accurately separate the tooth structure. The tooth feature quantitative extraction module further extracts multi-dimensional feature parameters from the segmentation results, including key indicators such as root length and curvature. Finally, the tooth extraction intelligent diagnosis module combines the clinical knowledge rule base to automatically give a tooth extraction difficulty assessment and main risk prediction conclusions based on the extracted feature parameters. The structured diagnosis report generation module integrates all relevant information to form a comprehensive diagnostic recommendation, thereby significantly improving the accuracy and efficiency of tooth extraction diagnosis.

[0072] This application also provides a method for automatically generating a tooth extraction diagnosis plan, such as Figure 5 As shown, the method for automatically generating a tooth extraction diagnosis plan includes: S1, acquiring CBCT image data of a patient object; S2, performing image preprocessing on the CBCT image data to obtain preprocessed CBCT image data; S3, performing refined segmentation based on boundary enhancement on the preprocessed CBCT image data to obtain a tooth image segmentation result map, wherein, in the process of refined segmentation based on boundary enhancement, regional boundaries are highlighted by constructing a directional sensitivity modulation map; S4, performing multi-dimensional feature extraction and quantification based on the tooth image segmentation result map to obtain tooth quantitative feature parameters; S5, inputting the tooth quantitative feature parameters into a pre-constructed clinical knowledge rule base to obtain a tooth extraction intelligent diagnosis conclusion; S6, integrating the tooth image segmentation result map, the tooth quantitative feature parameters and the tooth extraction intelligent diagnosis conclusion to obtain a structured diagnosis report.

[0073] The basic principles of this application have been described above with reference to specific examples. However, it should be noted that the advantages, strengths, and effects mentioned in this application are merely illustrative and not restrictive, and should not be considered as necessarily possessed by each example of this application. Furthermore, the specific details disclosed above are provided for illustrative purposes and to facilitate understanding, rather than as limitations. These details do not limit this application to necessarily being implemented using these specific details.

[0074] The block diagrams of the devices, devices, equipment, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, or configured in any manner. Words such as "include," "comprise," "have," and the like are open-ended words, meaning "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0075] It should also be noted that in the apparatus, device, and method of the present application, each component or each step can be decomposed and / or recombined, and such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0076] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0077] The above description has been provided for the purpose of illustration and description. Furthermore, this description is not intended to limit the examples of the present application to the forms disclosed herein. Although a number of example aspects and examples have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A system for automatically generating tooth extraction diagnosis plans, characterized in that: include: A CBCT image data acquisition module, used to acquire CBCT image data of a patient object; An image preprocessing module, configured to perform image preprocessing on the CBCT image data to obtain preprocessed CBCT image data; a dental image segmentation module, configured to perform boundary-enhanced refined segmentation on the pre-processed CBCT image data to obtain a dental image segmentation result map, wherein, during the boundary-enhanced refined segmentation process, a directional sensitivity modulation map is constructed to highlight regional boundaries; A tooth feature quantification extraction module, configured to extract and quantify multi-dimensional features based on the tooth image segmentation result image to obtain tooth quantitative feature parameters; A tooth extraction intelligent diagnosis module, configured to input the quantitative characteristic parameters of the tooth into a pre-built clinical knowledge rule base to obtain an intelligent diagnosis conclusion for tooth extraction; a structured diagnostic report generating module, configured to integrate the tooth image segmentation result map, the tooth quantitative characteristic parameters, and the tooth extraction intelligent diagnostic conclusion to obtain a structured diagnostic report; In the process of refined segmentation based on boundary enhancement, the regional boundaries are highlighted by constructing a directional sensitivity modulation map, including: Calculate the multiplication of the multi-level visual feature map of the tooth image and the boundary saliency prediction feature map by position point to obtain the initial tooth image boundary saliency visual feature map; Calculating a directional gradient amplitude feature map of the multi-level visual feature map of the tooth image relative to the boundary saliency prediction feature map; Calculating the partial derivatives of the eigenvalues ​​at each position in the directional gradient amplitude feature map relative to the eigenvalues ​​at each position in the multi-level visual feature map of the tooth image to obtain a first bidirectional partial derivative structure interaction coefficient modulation map; Calculating partial derivatives of the eigenvalues ​​at each position in the directional gradient amplitude feature map relative to the eigenvalues ​​at each position in the boundary saliency prediction feature map to obtain a second bidirectional partial derivative structure interaction coefficient modulation map; constructing a directional sensitivity modulation map based on the first bidirectional partial guide structure interaction coefficient modulation map and the second bidirectional partial guide structure interaction coefficient modulation map; The directional sensitivity modulation map and the initial tooth image boundary significant visual feature map are multiplied by position points to obtain the tooth image boundary significant visual feature map.

2. The automatic generation system of tooth extraction diagnosis plan according to claim 1, characterized in that: The image preprocessing module is used to perform standardization, denoising and contrast enhancement on the CBCT image data to obtain the preprocessed CBCT image data.

3. The automatic generation system of tooth extraction diagnosis plan according to claim 1, characterized in that: The dental image segmentation module includes: A dental image basic segmentation unit, configured to input the pre-processed CBCT image data into a basic segmentation network to obtain a dental image local visual feature map and a dental image global visual feature map; a dental image multi-level fusion unit, configured to fuse the dental image local visual feature map and the dental image global visual feature map to obtain a dental image multi-level visual feature map; A tooth image boundary saliency prediction unit, configured to input the tooth image multi-level visual feature map into a boundary saliency prediction network to obtain a boundary saliency prediction feature map; a tooth image boundary enhancement unit, configured to fuse the tooth image multi-level visual feature map and the boundary saliency prediction feature map to obtain a tooth image boundary enhancement visual feature map; The tooth image segmentation prediction unit is used to input the tooth image boundary enhancement visual feature map into the segmentation prediction layer to obtain the tooth image segmentation result map.

4. The automatic generation system of tooth extraction diagnosis plan according to claim 3, characterized in that: The basic segmentation network includes a hybrid model integrating CNN and Transformer.

5. The automatic generation system of tooth extraction diagnosis plan according to claim 3, characterized in that: The tooth image boundary saliency prediction unit is used to: Set the boundary segmentation threshold; All positions where the pixel values ​​in the multi-level visual feature map of the dental image are greater than the boundary segmentation threshold are marked as 1, and the remaining positions are marked as 0 to obtain the boundary significance prediction feature map.

6. The automatic generation system of tooth extraction diagnosis plan according to claim 3, characterized in that: The tooth image boundary strengthening unit includes: a boundary saliency fusion subunit, configured to fuse the multi-level visual feature map of the tooth image and the boundary saliency prediction feature map to obtain a boundary saliency visual feature map of the tooth image; The weighted calculation subunit is used to calculate the position-weighted sum between the multi-level visual feature map of the tooth image and the significant visual feature map of the tooth image boundary to obtain the tooth image boundary enhanced visual feature map.

7. The automatic generation system of tooth extraction diagnosis plan according to claim 1, characterized in that: The tooth feature quantitative extraction module includes: a tooth structure morphology measurement unit, configured to perform structure morphology measurement on the tooth image segmentation result image to obtain tooth root length, diameter, curvature, alveolar bone height, and bone density estimation values; a tooth spatial relationship calculation unit, configured to perform spatial relationship calculation on the tooth image segmentation result map to obtain the shortest distance between the tooth root tip and the inferior alveolar nerve canal or the maxillary sinus floor; The tooth lesion feature analysis unit is used to perform lesion feature analysis on the tooth image segmentation result image to obtain the volume, position and grayscale features of the apical lesion and cyst.

8. The automatic generation system of tooth extraction diagnosis plan according to claim 1, characterized in that: The tooth extraction intelligent diagnosis module is used to: The tooth quantitative characteristic parameters are input into a pre-built clinical knowledge rule base to obtain a tooth extraction difficulty assessment result and a risk prediction result, and the tooth extraction difficulty assessment result and the risk prediction result constitute the tooth extraction intelligent diagnosis conclusion.

9. A method for automatically generating a tooth extraction diagnosis plan, using the automatic generation system for a tooth extraction diagnosis plan according to any one of claims 1 to 8, characterized in that: include: Acquiring CBCT image data of a patient subject; performing image preprocessing on the CBCT image data to obtain preprocessed CBCT image data; Performing boundary-enhanced refined segmentation on the pre-processed CBCT image data to obtain a tooth image segmentation result map, wherein, in the process of boundary-enhanced refined segmentation, a directional sensitivity modulation map is constructed to highlight the regional boundaries; Performing multi-dimensional feature extraction and quantification based on the tooth image segmentation result image to obtain tooth quantitative feature parameters; Inputting the tooth quantitative characteristic parameters into a pre-built clinical knowledge rule base to obtain an intelligent diagnosis conclusion for tooth extraction; The tooth image segmentation result map, the tooth quantitative characteristic parameters and the tooth extraction intelligent diagnosis conclusion are integrated to obtain a structured diagnosis report.

Images