A feature enhancement system and method for breast and thyroid surgery CT images

Abstract

This application provides a feature enhancement system and method for breast and thyroid surgery CT images. First, a tissue recognition model is used to segment the target enhancement region of the target breast and thyroid surgery patient, and the CT image of the target region is extracted. Then, thermal imaging images of the target breast and thyroid surgery patient are acquired, and soft tissue identification boundaries and lesion connectivity regions of the target enhancement region are extracted based on the thermal imaging images. Soft tissue boundary enhancement is performed on the target region CT image according to the soft tissue identification boundaries to obtain an enhanced CT image, and image texture features of the enhanced CT image are extracted. Finally, consistency enhancement is performed on the image texture features of the enhanced CT image according to the lesion connectivity regions to obtain a visualized CT image of the target breast and thyroid surgery patient. This method can enhance breast and thyroid surgery CT images based on the soft tissue identification boundaries and lesion connectivity regions of the target patient's thermal imaging images, improving the texture coherence of the CT image in the lesion region.

Description

Technical Field

[0001] This application relates to the field of image feature enhancement technology, and more specifically, to a feature enhancement system and method for breast and thyroid surgery CT images. Background Technology

[0002] CT (Computed Tomography) is a commonly used medical imaging technique that uses X-ray beams to scan the human body layer by layer and reconstructs images using a computer, generating clear images of internal structures. Due to its advantages such as fast imaging speed and high detail resolution, CT is widely used in the diagnosis of clinical diseases. In the field of breast and thyroid surgery, CT images can be used to visually present the tissue structure and lesion characteristics of the breast and thyroid regions, assisting doctors in determining the specific location, shape, density, and spatial relationship of masses with surrounding tissues. This provides important information for differentiating between benign and malignant lesions, preoperative evaluation, and postoperative follow-up.

[0003] However, current technologies mainly rely on doctors' direct observation and experience in CT images for diagnosis. CT images have a certain degree of ambiguity when displaying soft tissue boundaries, and the texture features in key lesion areas are often discontinuous and indistinct, making identification difficult. This not only increases the reliance on doctors' professional experience, but also limits the further development and application of breast and thyroid surgery CT in high-precision diagnosis and intelligent analysis. Summary of the Invention

[0004] This application provides a feature enhancement system and method for breast and thyroid surgery CT images, which can enhance breast and thyroid surgery CT images based on the soft tissue identification boundaries and lesion connectivity regions of the target patient's thermal imaging images, thereby improving the texture coherence of CT images in the lesion region.

[0005] In a first aspect, this application provides a feature enhancement method for breast and thyroid surgery CT images. This method can be executed by a network device, or it can be executed by a chip configured in the network device. This application does not limit the method in this regard.

[0006] Specifically, the method includes:

[0007] Acquire initial CT images of the target breast and thyroid surgery patient, and preprocess the initial CT images;

[0008] The preprocessed initial CT image is acquired, and the target enhanced region of the target breast and thyroid surgery patient is segmented using a tissue recognition model to extract the CT image of the target region.

[0009] Acquire thermal imaging images of the target breast and thyroid surgical patient, and extract the soft tissue identification boundary and lesion connectivity region of the target enhanced region of the target breast and thyroid surgical patient based on the thermal imaging images;

[0010] Based on the soft tissue identification boundary, the CT image of the target region is enhanced with soft tissue boundary to obtain a CT enhanced image, and the image texture features of the CT enhanced image are extracted.

[0011] Based on the connected regions of the lesion, the image texture features of the enhanced CT image are uniformly enhanced to obtain the CT visualization image of the target breast and thyroid surgery patient.

[0012] In conjunction with the first aspect, in certain implementations of the first aspect, extracting the soft tissue identification boundary and lesion connectivity region of the target enhanced region of the target breast and thyroid surgical patient based on the thermal imaging image specifically includes:

[0013] The temperature matrix of the lesion is obtained by extracting the temperature matrix from the thermal imaging image.

[0014] Gradient features are extracted from the lesion temperature matrix using a temperature difference operator to obtain a lesion temperature gradient map.

[0015] Based on the temperature gradient distribution of the lesion temperature gradient map, a boundary segmentation threshold is determined, and based on the boundary segmentation threshold, boundary features are extracted from the lesion temperature gradient map to obtain the soft tissue identification boundary of the target enhancement region.

[0016] The region clustering algorithm is used to extract the connected regions of the lesion temperature gradient map, and the four-neighbor analysis algorithm is used to filter out the features of multiple connected regions of the lesion, retaining multiple connected regions with an area greater than a set threshold as the final connected regions of the lesion.

[0017] In conjunction with the first aspect, in certain implementations of the first aspect, performing soft tissue boundary enhancement on the CT image of the target region based on the soft tissue identification boundary to obtain a CT enhanced image specifically includes:

[0018] The soft tissue identification boundary of the thermal imaging image is obtained, and the soft tissue identification boundary is mapped to the CT image coordinate system of the target region CT image based on the spatial registration coordinate parameters to obtain the boundary mask in the CT image coordinate system.

[0019] Based on the boundary mask, local regions are extracted from the CT image. The boundary structure region covered by the mask is extracted, and the image grayscale features of the boundary structure region are obtained. The recognition boundary matching degree is determined based on the image grayscale features of the boundary structure region.

[0020] Based on the identified boundary matching degree, dynamic boundary enhancement is performed on the boundary structure region to obtain the CT enhanced image.

[0021] In conjunction with the first aspect, in some implementations of the first aspect, during the process of dynamically enhancing the boundary structure region based on the identification boundary matching degree, rule mapping is performed according to the identification boundary matching degree to determine the corresponding boundary enhancement parameters, and dynamic boundary enhancement operation is performed on the boundary structure region based on the boundary enhancement parameters.

[0022] In conjunction with the first aspect, in certain implementations of the first aspect, performing consistency enhancement on the image texture features of the CT-enhanced image based on the connected regions of the lesion to obtain a breast and thyroid surgery CT visualization image specifically includes:

[0023] Multiple connected regions of lesions in the thermal imaging image are obtained, and the multiple connected regions of lesions are mapped to the coordinate system of the CT enhanced image based on preset spatial registration parameters to form corresponding multiple connected region masks.

[0024] In CT enhanced images, for each connected region mask, the image texture features within the connected region mask are extracted, and the connected region matching degree corresponding to the connected region mask is determined based on the image texture features.

[0025] Based on the matching degree of each connected region mask, the pixel grayscale within each connected region mask is enhanced for consistency, resulting in a CT visualization image of breast and thyroid surgery.

[0026] In conjunction with the first aspect, in some implementations of the first aspect, a spiral CT scanner is used to acquire initial CT images of the target breast and thyroid surgical patient.

[0027] In conjunction with the first aspect, in some implementations of the first aspect, a convolutional neural network is used as a tissue recognition model to segment the target enhancement region of the target breast and thyroid surgical patient.

[0028] Secondly, this application provides a feature enhancement system for breast and thyroid surgery CT images, which includes an image processing unit, the image processing unit comprising:

[0029] The image acquisition module is used to acquire initial CT images of the target breast and thyroid surgery patient and to preprocess the initial CT images;

[0030] The image segmentation module is used to acquire the preprocessed initial CT image, and to segment the target enhancement region of the target breast and thyroid surgery patient using a tissue recognition model to extract the CT image of the target region.

[0031] The feature extraction module is used to acquire thermal imaging images of the target breast and thyroid surgical patient, and extract the soft tissue identification boundary and lesion connectivity region of the target enhancement area of ​​the target breast and thyroid surgical patient based on the thermal imaging images;

[0032] The image enhancement module is used to enhance the soft tissue boundary of the CT image of the target region based on the soft tissue identification boundary to obtain a CT enhanced image, and extract the image texture features of the CT enhanced image;

[0033] The image enhancement module is further configured to perform consistent enhancement of the image texture features of the CT enhanced image based on the connected regions of the lesion, thereby obtaining a CT visualization image of the target breast and thyroid surgery patient.

[0034] Thirdly, this application provides a computer terminal device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the aforementioned feature enhancement method for breast and thyroid surgical CT images.

[0035] Fourthly, this application provides a computer-readable storage medium storing at least one computer program, which is loaded and executed by a processor to perform the operations described above in the feature enhancement method for breast and thyroid surgical CT images.

[0036] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0037] This application provides a feature enhancement system and method for breast and thyroid surgery CT images. First, initial CT images of the target breast and thyroid surgery patient are acquired and preprocessed. The preprocessed initial CT images are then acquired, and a tissue recognition model is used to segment the target enhancement region of the patient, extracting the CT image of the target region. Next, thermal imaging images of the patient are acquired, and soft tissue identification boundaries and lesion connectivity regions of the target enhancement region are extracted based on the thermal imaging images. Soft tissue boundary enhancement is performed on the target region CT image based on the soft tissue identification boundaries to obtain an enhanced CT image, and image texture features of the enhanced CT image are extracted. Finally, consistency enhancement is performed on the image texture features of the enhanced CT image based on the lesion connectivity regions to obtain a visualized CT image of the target breast and thyroid surgery patient.

[0038] Therefore, it can be seen that this application can reflect tissue metabolic activity and blood flow changes by capturing the temperature distribution of the tissue surface through infrared thermal imaging. It uses the temperature gradient abrupt change at the boundary to correspond to the natural boundary of the tissue structure. Through the thermal anomaly accompanying the lesion area (such as nodules and masses), a stable connected region is formed. Based on the soft tissue boundary and lesion connected region extracted by thermal imaging, it can make up for the problems of unclear gray-scale transition in soft tissue and insufficient expression of weak texture area at the lesion boundary in CT images. Furthermore, it uses thermal imaging boundary to guide CT edge enhancement, improves the gray-scale gradient at the soft tissue junction, and solves the problem of blurred soft tissue boundary. In the lesion connected region, the consistency of thermal texture is constrained to ensure the overall coherence of the texture and the uniformity of local details in the region.

[0039] In summary, this application, by introducing soft tissue identification boundaries and lesion connectivity regions from thermal imaging images, achieves structural guidance and enhanced regional consistency in CT images, compensating for the inherent weaknesses of CT in soft tissue imaging, significantly improving the texture coherence and visualization effect of lesion regions, thereby enhancing the image diagnosis accuracy and reliability of breast and thyroid surgical diseases. Attached Figure Description

[0040] Figure 1 This is an exemplary flowchart of a method for feature enhancement of breast and thyroid surgical CT images according to some embodiments of this application;

[0041] Figure 2 This is a schematic diagram of the structure of an image processing unit according to some embodiments of this application;

[0042] Figure 3 This is a schematic diagram of the structure of a computer terminal device that implements a feature enhancement method for breast and thyroid surgical CT images according to some embodiments of this application. Detailed Implementation

[0043] This application acquires initial CT images of a target breast and thyroid surgical patient, preprocesses the initial CT images, obtains the preprocessed initial CT images, segments the target enhancement region of the target breast and thyroid surgical patient using a tissue recognition model, and extracts the CT image of the target region. It then acquires thermal imaging images of the target breast and thyroid surgical patient, extracts the soft tissue identification boundary and lesion connectivity region of the target enhancement region based on the thermal imaging images, enhances the soft tissue boundary of the target region CT image according to the soft tissue identification boundary, obtains the CT enhanced image, and extracts the image texture features of the CT enhanced image. Finally, it performs consistency enhancement on the image texture features of the CT enhanced image based on the lesion connectivity region, obtaining the CT visualization image of the target breast and thyroid surgical patient. This method can enhance the CT image of breast and thyroid surgery based on the soft tissue identification boundary and lesion connectivity region of the target patient's thermal imaging image, improving the texture coherence of the CT image in the lesion region.

[0044] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific implementation methods. (Reference) Figure 1 The figure is an exemplary flowchart of a feature enhancement method for breast and thyroid surgical CT images according to some embodiments of this application. The feature enhancement method 100 for breast and thyroid surgical CT images mainly includes the following steps:

[0045] In step S101, initial CT images of the target breast and thyroid surgery patient are acquired, and the initial CT images are preprocessed.

[0046] Preferably, in some embodiments, a spiral CT scanner is used to acquire the initial CT images of the target breast and thyroid surgery patient. In some other embodiments, other devices or equipment capable of acquiring CT images may also be used, and this application does not limit this.

[0047] Optionally, in some embodiments, the slice thickness of the initial CT image is set to 0.1 mm, the scanning position of the initial CT image is from the mandible to the lower edge of the chest of the breast and thyroid surgery patient, and whether to inject contrast agent for enhancement is determined as needed.

[0048] Optionally, in some embodiments, preprocessing the initial CT image includes denoising and grayscale normalization of the initial CT image to improve image quality and enhance the accuracy of subsequent image processing and fusion.

[0049] In specific implementation, Gaussian filtering can be used to denoise the initial CT image. By weighted summation of each pixel and its neighborhood, speckle noise can be effectively reduced, improving image smoothness and edge preservation. In some embodiments, nonlocal mean filtering can also be used to perform detail-preserving denoising, thereby preserving the texture information in the thyroid and breast tissues.

[0050] In the grayscale normalization process of preprocessing the initial CT image, the grayscale values ​​of the CT image can be normalized to the range of [0, 255] using a linear normalization method to reduce the impact of image brightness differences between devices. In other embodiments, the Z-score normalization method can also be used to normalize the standard deviation of the image pixel intensity.

[0051] Furthermore, in some preferred embodiments, the preprocessing further includes spatial registration of the initial CT image and the thermal image, specifically including a rigid affine or non-rigid B-spline registration algorithm, through key point matching and transformation matrix calculation, to achieve accurate spatial alignment of different modal images and obtain the corresponding spatial registration parameters. For example, the transformation matrix of the spatial coordinate system is used as the spatial registration parameter to ensure that the thermal imaging information can be mapped to the corresponding CT anatomical structure region.

[0052] In terms of image resolution during the preprocessing of the initial CT image, to ensure the uniformity of subsequent model processing, in some embodiments of this application, the preprocessed CT image may be resampled to standardize the pixel spacing to a preset voxel spacing (e.g., 1.0mm × 1.0mm × 1.0mm), and bilinear interpolation or cubic spline interpolation may be used to reconstruct the image to improve the consistency of image processing for different patients.

[0053] In step S102, the preprocessed initial CT image is acquired, and the target enhanced region of the target breast and thyroid surgery patient is segmented using a tissue recognition model to extract the CT image of the target region.

[0054] It should be noted that the target enhancement region refers to a specific image region in the breast and thyroid surgery image that has been selected for subsequent image processing after recognition model or rule analysis. In specific implementation, a fixed scanning range can be selected as the target enhancement region based on the scanning position of the initial CT image and the image segmentation can be performed. Alternatively, after obtaining the preprocessed initial CT image, a tissue recognition model can be used to segment the target enhancement region of the target breast and thyroid surgery patient to extract the corresponding thyroid or breast region CT image.

[0055] In specific implementation, a tissue recognition model can be constructed based on the convolutional neural network structure commonly used in the field of medical image segmentation. Preferably, in some embodiments, the tissue recognition model is a deep segmentation network based on the U-Net architecture, which has skip connections and an encoder-decoder structure, and can effectively identify soft tissue boundaries and lesion areas in CT images.

[0056] When training the tissue recognition model, a professionally labeled CT image dataset of breast and thyroid surgery can be used. The training samples contain pixel-level masks of the thyroid gland, breast, and their typical lesion areas to optimize the model in a supervised manner. Preferably, the Dice loss function or cross-entropy loss function is introduced during the training process to improve the segmentation accuracy.

[0057] In some embodiments, to enhance the model's adaptability to different patients and image modalities, the tissue recognition model may also introduce attention mechanisms, multi-scale feature extraction structures (such as ASPP modules), or multi-branch fusion network structures to improve the model's ability to recognize regions with blurred boundaries and similar densities.

[0058] In the actual reasoning process, the preprocessed initial CT image is input into the trained tissue recognition model, and a binary segmentation image of the target enhanced region is output. The foreground region in the binary image corresponds to the thyroid and breast surgical anatomical structure (such as thyroid tissue, breast parenchyma or lesion). Subsequently, the corresponding target region CT image is extracted from the original CT image based on the mask of this region.

[0059] Optionally, to further improve segmentation accuracy, in some embodiments, the tissue recognition model can also be used in conjunction with traditional medical image segmentation algorithms such as graph cut and region growing to form a combined strategy of coarse segmentation and fine repair, thereby achieving higher quality image region extraction.

[0060] In step S103, thermal imaging images of the target breast and thyroid surgical patient are acquired, and soft tissue identification boundaries and lesion connectivity regions of the target enhanced region of the target breast and thyroid surgical patient are extracted based on the thermal imaging images.

[0061] It should be noted that infrared thermography, by capturing the temperature distribution on the tissue surface, can reflect changes in tissue metabolic activity and blood flow, exhibiting highly sensitive temperature contrast characteristics. In breast and thyroid surgery, tissue boundaries often appear as abrupt regions of temperature gradient, while lesion areas (such as nodules and masses) are accompanied by stable thermal anomalies, forming a clearly connected structure. Therefore, thermography images can effectively extract soft tissue boundaries and lesion connectivity areas, compensating for the shortcomings of CT images in terms of blurred soft tissue grayscale transitions and insufficient lesion texture representation, thereby improving the structural visibility and diagnostic accuracy of lesion areas.

[0062] Optionally, in some embodiments, thermal imaging images of the target breast and thyroid surgical patient are acquired using an infrared thermal imaging camera to obtain temperature distribution information of the body surface and superficial tissues. The enhanced areas of the target are further identified and analyzed in conjunction with the thermal imaging images. Specifically, the resolution of the thermal imaging images is 640×480 or higher, the image grayscale values ​​correspond to different temperature gradients, and the color mapping can adopt the JET pseudo-color mapping method to enhance the visual expression of temperature differences.

[0063] Optionally, in some embodiments, before extracting the soft tissue identification boundary and lesion connectivity region of the target enhanced region of the target breast and thyroid surgical patient based on the thermal imaging image, the method further includes: spatial registration of the initial CT image and the thermal imaging image, specifically including a rigid affine or non-rigid B-spline registration algorithm, through key point matching and transformation matrix calculation, to achieve precise spatial alignment of the initial CT image and the thermal imaging image, thereby ensuring that the thermal imaging information is mapped to the corresponding CT anatomical structure region.

[0064] Preferably, in some embodiments, extracting the soft tissue identification boundary and lesion connectivity region of the target enhanced region of the target breast and thyroid surgical patient based on the thermal imaging image specifically includes:

[0065] The temperature matrix of the lesion is obtained by extracting the temperature matrix from the thermal imaging image.

[0066] Gradient features are extracted from the lesion temperature matrix using a temperature difference operator to obtain a lesion temperature gradient map.

[0067] Based on the temperature gradient distribution of the lesion temperature gradient map, a boundary segmentation threshold is determined, and based on the boundary segmentation threshold, boundary features are extracted from the lesion temperature gradient map to obtain the soft tissue identification boundary of the target enhancement region.

[0068] The region clustering algorithm is used to extract the connected regions of the lesion temperature gradient map, and the four-neighbor analysis algorithm is used to filter out the features of multiple connected regions of the lesion, retaining multiple connected regions with an area greater than a set threshold as the final connected regions of the lesion.

[0069] In practice, the temperature matrix extraction can be based on the pixel grayscale values ​​of thermal imaging images to perform temperature inversion. By using a lookup table or a calibration model of the thermal imaging device, the grayscale value of each pixel is converted into the corresponding temperature value, and a two-dimensional lesion temperature matrix is ​​constructed to reflect the actual temperature distribution in the breast and thyroid surgery-related areas.

[0070] Furthermore, in the process of extracting temperature gradient features, it is preferable to use the Sobel operator or a custom temperature difference convolution kernel to perform local gradient calculation on the lesion temperature matrix, extract the edge features of the temperature change region, construct the lesion temperature gradient map, and thus identify the soft tissue boundary and the region of abnormal thermal change.

[0071] In some implementations, the boundary segmentation threshold can be determined using an adaptive method, specifically including: calculating the global mean, standard deviation, or Gaussian distribution characteristics in the temperature gradient map, and setting a dynamic threshold related to local background contrast to enhance adaptability among different patients. In some other implementations, histogram analysis can be performed on the temperature gradient map G(x,y), and the boundary segmentation threshold can be set based on the histogram, such as selecting the mean or using Otsu's automatic segmentation threshold. Based on this threshold, a set of all pixels whose temperature gradient exceeds the boundary segmentation threshold is extracted to form a preliminary boundary mask. Alternatively, Otsu's automatic thresholding method can be used to binarize the gradient map and extract edge candidate regions. Finally, the coherence and closure of the soft tissue recognition boundary are further optimized through edge tracking and morphological operations (such as closing operations and erosion) to obtain a clearly structured target enhancement region soft tissue recognition boundary. Specifically, morphological dilation and closing operations are performed on the preliminary boundary mask to remove isolated noise points and broken boundaries; then, contour detection (such as based on connected components or Canny edge tracking) is combined to extract coherent curves, ultimately generating the soft tissue recognition boundary of the target enhancement region.

[0072] It should be noted that this invention can extract soft tissue boundaries for breast and thyroid surgery from thermal imaging images. This is primarily based on the significant differences in metabolic activity and blood perfusion in soft tissues, which manifest as localized temperature abrupt changes in the thermal image, particularly at tissue boundaries where a clear temperature gradient forms, constituting a natural thermal boundary. By performing gradient calculations on the temperature matrix, the intensity of temperature changes in edge regions can be effectively identified, generating a temperature gradient map. Combined with boundary segmentation thresholds and connected component analysis, the location of tissue boundaries can be accurately pinpointed. This method achieves stable extraction of physiologically significant soft tissue structural boundaries from thermal images, providing a reliable structural prior for subsequent enhancement of CT images.

[0073] During the extraction of the connected regions of the lesions, K-means, Mean-Shift, or density-based spatial clustering algorithms (such as DBSCAN) are used to perform temperature similarity clustering on the temperature gradient map of the lesions, and multiple clustered regions with local temperature consistency are initially divided.

[0074] Furthermore, connectivity is assessed based on a four-neighborhood analysis algorithm (or an eight-neighborhood analysis algorithm) of the preliminary clustering results to identify the connected regions that actually constitute the anatomical structure. In some embodiments, a minimum area threshold can be set (e.g., greater than 100 pixels or an actual area greater than 5 mm). 2 After removing isolated points, small patches, and misjudged background areas, only multiple connected blocks with continuous structure and reasonable morphology are retained as the final target lesion connected regions for enhancement.

[0075] Optionally, in some preferred embodiments, the spatial relationship between the soft tissue identification boundary and the lesion connected region can be logically verified, such as determining whether the boundary is closed and whether the connected region falls within the boundary range. A topology correction algorithm can also be introduced to further improve the accuracy of region extraction and the consistency of anatomical structure.

[0076] Preferably, in some embodiments, after extracting the soft tissue identification boundary and lesion connectivity region of the target enhanced region of the target breast and thyroid surgical patient based on the thermal imaging image, the method further includes: mapping the soft tissue identification boundary and the lesion connectivity region to the CT image coordinate system based on the spatial registration relationship between the completed thermal imaging image and the initial CT image.

[0077] In step S104, the soft tissue boundary of the target region CT image is enhanced according to the soft tissue identification boundary to obtain a CT enhanced image, and the image texture features of the CT enhanced image are extracted.

[0078] Optionally, in some embodiments, performing soft tissue boundary enhancement on the CT image of the target region based on the soft tissue identification boundary to obtain a CT enhanced image specifically includes:

[0079] The soft tissue identification boundary of the thermal imaging image is obtained, and the soft tissue identification boundary is mapped to the CT image coordinate system of the target region CT image based on the spatial registration coordinate parameters to obtain the boundary mask in the CT image coordinate system.

[0080] Based on the boundary mask, local regions are extracted from the CT image. The boundary structure region covered by the mask is extracted, and the image grayscale features of the boundary structure region are obtained. The recognition boundary matching degree is determined based on the image grayscale features of the boundary structure region.

[0081] Based on the identified boundary matching degree, dynamic boundary enhancement is performed on the boundary structure region to obtain the CT enhanced image.

[0082] It should be noted that this scheme spatially registers thermal and CT images, mapping the boundary and lesion region information in the thermal image to the CT image coordinate system to generate a structural constraint mask. The boundary mask is used to limit the local enhancement range and prevent blind processing; the lesion connectivity mask provides a reference for regional coherence and avoids texture breaks. These masks, as structural priors, guide the enhancement process to be executed in a targeted manner, thereby improving the accuracy and structural consistency of image enhancement.

[0083] In specific implementation, the soft tissue boundary enhancement of the target region CT image is performed based on the soft tissue identification boundary to obtain the CT enhanced image, specifically including:

[0084] First, the soft tissue identification boundary extracted from the thermal imaging image is obtained, and based on the previously established spatial registration coordinate parameters, the soft tissue identification boundary is mapped to the CT image coordinate system of the target area CT image to obtain the corresponding boundary position and its spatial coordinate contour.

[0085] Furthermore, a boundary mask in the CT image coordinate system is generated based on the mapped soft tissue identification boundary. This mask is used to indicate the structural boundary regions in the target CT image that should be given priority processing.

[0086] Next, the boundary mask is used to extract local regions of the CT image of the target area, and the boundary structure region covered by the mask is extracted. Based on this, the image grayscale features of the boundary structure region are obtained, including but not limited to the average grayscale value, edge gradient intensity distribution, texture contrast, etc.

[0087] In some implementations, the identification boundary matching degree of the boundary structure region is further calculated, which is defined as the average value of the overall gray-level gradient of the region in the direction of the boundary normal (i.e., vertical). The identification boundary matching degree mentioned in this application is used to measure the degree of alignment between the infrared boundary position and the actual structural boundary in the CT image. Based on the identification boundary matching degree, dynamic boundary enhancement is performed on the boundary structure region of the target region CT image. In this way, the influence of the enhanced direction filter output is strengthened in the low matching degree region, thereby enhancing the edge response of the image. It should be noted that the edge response mentioned in this application can be equivalently replaced by the gray-level change rate of the image at the edge of the selected structural region. This application will not elaborate further on this.

[0088] Preferably, in some embodiments, during the process of dynamically enhancing the boundary structure region based on the identified boundary matching degree, rule mapping can be performed according to the identified boundary matching degree to determine the corresponding boundary enhancement parameters, and dynamic boundary enhancement operation can be performed on the boundary structure region based on the boundary enhancement parameters. Specifically, the boundary enhancement parameters can be determined according to the following formula:

[0089]

[0090] in The boundary enhancement parameters, Here, exp is a predefined maximum boundary enhancement parameter, and exp is a logarithmic function. The response coefficient, based on experience, can be set in the range of 0.05–0.2. Based on the boundary matching degree, a directional filter can be used to enhance the edge response in the normal direction based on the boundary enhancement parameter. The boundary enhancement parameter is used to proportionally enhance the edge information of the directional filter. For example, a Sobel directional filter is used to extract the vertical edge response of the boundary structure region as R(x,y), and the boundary enhancement parameter is 1.2. The final enhanced image result is the overlap image of the initial target region CT image and the vertical edge response multiplied by 1.2.

[0091] In step S105, the image texture features of the enhanced CT image are uniformly enhanced according to the connected regions of the lesion to obtain a breast and thyroid surgery CT visualization image.

[0092] It should be noted that this application proposes a two-level synergistic enhancement mechanism of "boundary enhancement + texture consistency enhancement". By guiding the edge enhancement of CT images through thermal imaging boundaries, the gray-scale gradient at the soft tissue junction is effectively improved, the boundary blurring problem is alleviated, and the structural recognition capability is enhanced. At the same time, the consistency characteristics of the thermal image are used in the lesion connected area to perform gradient normalization, histogram standardization or low-frequency texture compensation in the CT image of areas with weak or discontinuous texture, so as to achieve the overall coherence and detail uniformity of texture in the region, thereby ensuring clear image structure and stable and reliable lesion expression.

[0093] Preferably, in some embodiments, performing consistency enhancement on the image texture features of the CT-enhanced image based on the connected regions of the lesion to obtain a breast and thyroid surgery CT visualization image specifically includes:

[0094] Multiple connected regions of lesions in the thermal imaging image are obtained, and the multiple connected regions of lesions are mapped to the coordinate system of the CT enhanced image based on preset spatial registration parameters to form corresponding multiple connected region masks.

[0095] In CT enhanced images, for each connected region mask, the image texture features within the connected region mask are extracted, and the connected region matching degree corresponding to the connected region mask is determined based on the image texture features.

[0096] Based on the matching degree of each connected region mask, the pixel grayscale within each connected region mask is enhanced for consistency, resulting in a CT visualization image of breast and thyroid surgery.

[0097] In specific implementation, firstly, multiple lesion connected regions in the thermal imaging image are obtained, and based on the established spatial registration parameters (such as affine transformation matrix or nonlinear B-spline mesh), the multiple lesion connected regions are mapped to the coordinate system of the CT enhanced image to form corresponding multiple connected region masks, which are used to indicate the structural position of thermal imaging information in the CT image.

[0098] Subsequently, in the CT-enhanced image, for each connected region mask, image texture features within that mask are extracted. Preferably, the image texture features include, but are not limited to: the average grayscale value and standard deviation of pixels within the region, gradient histogram distribution, grayscale co-occurrence matrix, etc., and then, based on the image texture features, the connected region matching degree corresponding to the connected region mask is calculated. Preferably, the matching degree is defined as the ratio of the average grayscale gradient change ∇I in the infrared thermal imaging image to the average grayscale gradient change ∇CT in the CT-enhanced image, and the connected region matching degree is used to reflect the image texture clarity of different connected region masks.

[0099] Preferably, in some embodiments, during the process of enhancing the consistency of pixel grayscale within each connected region mask based on the connected region matching degree corresponding to each connected region mask, the pixel grayscale within the connected region mask can be determined according to the following formula:

[0100]

[0101] in, The grayscale value of the pixel at (x, y) coordinates within the connected region mask after consistency enhancement. The average grayscale value of the image within the connected region mask. The degree of matching of the connected region. To ensure consistency enhancement, the grayscale value of the pixel at the (x,y) coordinate within the connected region mask is... The standard deviation of image grayscale within the connected region mask is given. The preset standard deviation of the target image grayscale.

[0102] This application introduces soft tissue identification boundaries and lesion connectivity regions from thermal imaging images to achieve structural guidance and enhanced regional consistency in CT images. This compensates for the inherent weaknesses of CT in soft tissue imaging, significantly improves the texture coherence and visualization effect of lesion regions, and thus enhances the accuracy and reliability of image diagnosis for breast and thyroid diseases.

[0103] Furthermore, in another aspect of this application, in some embodiments, this application provides a feature enhancement system for breast and thyroid surgery CT images, the system including an image processing unit, referenced... Figure 2 The figure is a schematic diagram of the exemplary hardware and / or software structure of an image processing unit according to some embodiments of this application. The image processing unit 200 includes: an image acquisition module 201, an image segmentation module 202, a feature extraction module 203, and an image enhancement module 204, which are described below:

[0104] Image acquisition module 201 is used to acquire initial CT images of the target breast and thyroid surgery patient and to preprocess the initial CT images;

[0105] Image segmentation module 202 is used to acquire the preprocessed initial CT image, segment the target enhancement region of the target breast and thyroid surgery patient using a tissue recognition model, and extract the CT image of the target region.

[0106] Feature extraction module 203 is used to acquire thermal imaging images of the target breast and thyroid surgical patient, and extract the soft tissue identification boundary and lesion connectivity region of the target enhanced region of the target breast and thyroid surgical patient based on the thermal imaging images;

[0107] Image enhancement module 204 is used to enhance the soft tissue boundary of the CT image of the target region based on the soft tissue identification boundary to obtain a CT enhanced image and extract the image texture features of the CT enhanced image;

[0108] The image enhancement module 204 is further configured to perform consistent enhancement of the image texture features of the CT enhanced image based on the connected regions of the lesion, thereby obtaining a CT visualization image of the target breast and thyroid surgical patient. The foregoing has detailed an example of a feature enhancement system and method for breast and thyroid surgical CT images provided by embodiments of this application. It is understood that, in order to achieve the above functions, the corresponding device includes hardware structures and / or software modules corresponding to the execution of each function.

[0109] Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a certain function in the application is executed in a manner that drives hardware or computer software depends on the specific application and design constraints of the technical solution. Therefore, those skilled in the art can use different methods to implement the described function for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0110] In addition, this application also provides a computer terminal device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described feature enhancement method for breast and thyroid surgical CT images.

[0111] In some embodiments, reference Figure 3 The figure is a schematic diagram of the structure of a computer terminal device implementing a feature enhancement method for breast and thyroid surgical CT images according to some embodiments of this application. The feature enhancement method for breast and thyroid surgical CT images described in the above embodiments can be achieved through... Figure 3The computer terminal device 300 shown is used to implement this, and the computer terminal device 300 includes at least one communication bus 301, communication interface 302, processor 303 and memory 304.

[0112] The processor 303 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more devices used to control the execution of a feature enhancement method for a breast and thyroid surgical CT image in this application.

[0113] The communication bus 301 may include a path for transmitting information between the aforementioned components.

[0114] Memory 304 may be read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions, random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 304 may exist independently and be connected to processor 303 via communication bus 301. Memory 304 may also be integrated with processor 303.

[0115] The memory 304 stores program code for executing the scheme of this application, and its execution is controlled by the processor 303. The processor 303 executes the program code stored in the memory 304. The program code may include one or more software modules. In the above embodiments, the determination of the soft tissue identification boundary can be achieved by the processor 303 and one or more software modules in the program code in the memory 304.

[0116] Communication interface 302 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0117] Optionally, the computer terminal device 300 may also include a power supply 305 for providing power to various devices or circuits in the real-time computer terminal device.

[0118] In a specific implementation, as one example, a computer terminal device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0119] The aforementioned computer terminal device can be a general-purpose computer terminal device or a dedicated computer terminal device. In specific implementations, the computer terminal device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer terminal device.

[0120] In addition, other aspects of this application provide a computer-readable storage medium storing at least one computer program that is loaded and executed by a processor to perform the operations performed by the above-described feature enhancement method for breast and thyroid surgical CT images.

[0121] In summary, the feature enhancement system and method for breast and thyroid surgery CT images disclosed in this application first acquires an initial CT image of the target breast and thyroid surgery patient and preprocesses the initial CT image; after acquiring the preprocessed initial CT image, a tissue recognition model is used to segment the target enhancement region of the target breast and thyroid surgery patient, and the CT image of the target region is extracted; a thermal imaging image of the target breast and thyroid surgery patient is acquired, and the soft tissue recognition boundary and lesion connectivity region of the target enhancement region of the target breast and thyroid surgery patient are extracted based on the thermal imaging image; soft tissue boundary enhancement is performed on the CT image of the target region based on the soft tissue recognition boundary to obtain a CT enhanced image, and the image texture features of the CT enhanced image are extracted; the image texture features of the CT enhanced image are uniformly enhanced based on the lesion connectivity region to obtain a CT visualization image of the target breast and thyroid surgery patient. This system can enhance the CT image of breast and thyroid surgery based on the soft tissue recognition boundary and lesion connectivity region of the target patient's thermal imaging image, thereby improving the texture coherence of the CT image in the lesion region.

[0122] The above descriptions are merely embodiments of this application, and common knowledge such as specific technical solutions or characteristics in the solutions are not described in detail here. It should be noted that those skilled in the art can make several modifications and improvements without departing from the technical solutions of this application, and these should also be considered within the scope of protection of this application, without affecting the effectiveness of the implementation of this application or the practicality of the patent.

[0123] The scope of protection claimed in this application shall be determined by the content of its claims. The specific embodiments described in the specification can be used to interpret the content of the claims. Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of the invention. Therefore, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A feature enhancement method for breast and thyroid surgical CT images, characterized in that, include: Acquire initial CT images of the target breast and thyroid surgery patient, and preprocess the initial CT images; The preprocessed initial CT image is acquired, and the target enhanced region of the target breast and thyroid surgery patient is segmented using a tissue recognition model to extract the CT image of the target region. Acquire thermal imaging images of the target breast and thyroid surgical patient, and extract the soft tissue identification boundary and lesion connectivity region of the target enhanced region of the target breast and thyroid surgical patient based on the thermal imaging images; Based on the soft tissue identification boundary, the CT image of the target region is enhanced with soft tissue boundary to obtain a CT enhanced image, and the image texture features of the CT enhanced image are extracted. Based on the connected regions of the lesions, the image texture features of the enhanced CT image are uniformly enhanced to obtain the CT visualization image of the target breast and thyroid surgery patient; Specifically, extracting the soft tissue identification boundary and lesion connectivity region of the target enhanced region of the target breast and thyroid surgical patient based on the thermal imaging image includes: The temperature matrix of the lesion is obtained by extracting the temperature matrix from the thermal imaging image. Gradient features are extracted from the lesion temperature matrix using a temperature difference operator to obtain a lesion temperature gradient map. Based on the temperature gradient distribution of the lesion temperature gradient map, a boundary segmentation threshold is determined, and based on the boundary segmentation threshold, boundary features are extracted from the lesion temperature gradient map to obtain the soft tissue identification boundary of the target enhancement region. The region clustering algorithm is used to extract the connected regions of the lesion temperature gradient map, and the four-neighbor analysis algorithm is used to filter out the features of multiple connected regions of the lesion, retaining multiple connected regions with an area greater than a set threshold as the final connected regions of the lesion. Based on the soft tissue identification boundary, the CT image of the target region is enhanced with soft tissue boundary information to obtain an enhanced CT image. Specifically, this includes: The soft tissue identification boundary of the thermal imaging image is obtained, and the soft tissue identification boundary is mapped to the CT image coordinate system of the target region CT image based on the spatial registration coordinate parameters to obtain the boundary mask in the CT image coordinate system. Based on the boundary mask, local regions are extracted from the CT image. The boundary structure region covered by the mask is extracted, and the image grayscale features of the boundary structure region are obtained. The recognition boundary matching degree is determined based on the image grayscale features of the boundary structure region. Based on the identified boundary matching degree, dynamic boundary enhancement is performed on the boundary structure region to obtain the CT enhanced image.

2. The method as described in claim 1, characterized in that, During the process of dynamically enhancing the boundary structure region based on the identification boundary matching degree, rule mapping is performed according to the identification boundary matching degree to determine the corresponding boundary enhancement parameters, and dynamic boundary enhancement operation is performed on the boundary structure region based on the boundary enhancement parameters.

3. The method as described in claim 1, characterized in that, Based on the connected regions of the lesions, the image texture features of the enhanced CT image are uniformly enhanced to obtain a breast and thyroid surgery CT visualization image, specifically including: Multiple connected regions of lesions in the thermal imaging image are obtained, and the multiple connected regions of lesions are mapped to the coordinate system of the CT enhanced image based on preset spatial registration parameters to form corresponding multiple connected region masks. In CT enhanced images, for each connected region mask, the image texture features within the connected region mask are extracted, and the connected region matching degree corresponding to the connected region mask is determined based on the image texture features. Based on the matching degree of each connected region mask, the pixel grayscale within each connected region mask is enhanced for consistency, resulting in a CT visualization image of breast and thyroid surgery.

4. The method as described in claim 1, characterized in that, Initial CT images of the target breast and thyroid surgery patients were acquired using a spiral CT scanner.

5. The method as described in claim 1, characterized in that, A convolutional neural network was used as a tissue recognition model to segment the target enhanced region of the target breast and thyroid surgery patient.

6. A feature enhancement system for breast and thyroid surgical CT images, comprising an image processing unit, said image processing unit being used to execute the feature enhancement method for breast and thyroid surgical CT images according to any one of claims 1 to 5, characterized in that, The image processing unit includes: The image acquisition module is used to acquire initial CT images of the target breast and thyroid surgery patient and to preprocess the initial CT images; The image segmentation module is used to acquire the preprocessed initial CT image, and to segment the target enhancement region of the target breast and thyroid surgery patient using a tissue recognition model to extract the CT image of the target region. The feature extraction module is used to acquire thermal imaging images of the target breast and thyroid surgical patient, and extract the soft tissue identification boundary and lesion connectivity region of the target enhancement area of ​​the target breast and thyroid surgical patient based on the thermal imaging images; The image enhancement module is used to enhance the soft tissue boundary of the CT image of the target region based on the soft tissue identification boundary to obtain a CT enhanced image, and extract the image texture features of the CT enhanced image; The image enhancement module is further configured to perform consistent enhancement of the image texture features of the CT enhanced image based on the connected regions of the lesion, thereby obtaining a CT visualization image of the target breast and thyroid surgery patient.

7. A computer terminal device, characterized in that, The computer terminal device includes a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute a feature enhancement method for breast and thyroid surgical CT images as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing at least one computer program, characterized in that, The computer program is loaded and executed by a processor to perform the operations of a feature enhancement method for breast and thyroid surgical CT images as described in any one of claims 1 to 5.

Images