AI assistance-based lung cancer pleural invasion diagnosis auxiliary method and system

Abstract

The invention belongs to the field of pleural invasion diagnosis, and particularly relates to a lung cancer pleural invasion diagnosis auxiliary method and system based on AI assistance, and the method comprises the steps: firstly obtaining a scanning image and an electrode needle signal, and obtaining an enhanced training set through time synchronization and enhanced labeling; secondly, constructing a regional prediction model based on a YOLOv8 algorithm and an attention matching network, and training the model by using an enhanced training set and a regional electric signal conversion error function; and 3, inputting a real-time scanning image into the trained region prediction model to obtain a suspected probing region and an electrode needle probing prediction signal sequence, configuring a safe probing boundary, combining a real-time electrode needle signal and a prediction signal to obtain a probing signal error, inputting the probing signal error into a probing path correction model for training, obtaining an optimal probing path through risk assessment, and finally, obtaining a safe probing result. And integrating the two models to an electrode needle, assisting in correcting an electrode needle probing path according to real-time data, and obtaining a probing result and an invasion probing score of the suspected probing area.

Description

Technical Field

[0001] The present invention belongs to the field of pleural invasion diagnosis, and in particular to an AI-assisted lung cancer pleural invasion diagnosis auxiliary method and system. Background Art

[0002] In the diagnosis and treatment of lung cancer, accurate judgment of pleural invasion is crucial for formulating a reasonable treatment plan. Traditionally, doctors rely on imaging examinations (such as CT, MRI) and pathological biopsies to assess whether the pleura is invaded. However, these methods have many limitations. Currently, diagnosis relies only on a single scan image or electrode needle signal, and information acquisition is incomplete, which can easily lead to misdiagnosis or missed diagnosis. In addition, when dealing with the complex relationship between lung structure and lesions, existing technologies are difficult to accurately locate and assess pleural invasion. In addition, manual interpretation of image and signal data is not only time-consuming and labor-intensive, but also affects diagnostic accuracy due to individual experience differences.

[0003] For example, the Chinese patent with the authorization announcement number CN116739992B discloses an intelligent auxiliary interpretation method for thyroid capsule invasion, which mainly adopts a biomedical image segmentation method that provides efficient local-global fusion and Transformer and CNN parallel networks. This prior art can output a positioning and segmentation image of suspected thyroid nodules by inputting a thyroid ultrasound image, and automatically obtain the preliminary interpretation result of thyroid nodules.

[0004] For example, a Chinese patent application with publication number CN115700761A discloses a method, device and electronic device for predicting microvascular invasion of hepatocellular carcinoma. The method includes: obtaining preoperative imaging images, clinical indicators and corresponding lesion pathology results of treated patients; training a pre-constructed prediction model based on the preoperative imaging images, clinical indicators and corresponding lesion pathology results of treated patients, and determining that the prediction model is a microvascular invasion prediction model when the prediction model meets the preset standards; obtaining preoperative imaging images and clinical indicators of the patient to be predicted; inputting the preoperative imaging images and clinical indicators of the patient to be predicted into the microvascular invasion prediction model, and outputting a classification result to predict whether the patient to be tested has microvascular invasion of hepatocellular carcinoma.

[0005] The above existing technologies have the following problems: the existing technologies mostly use a single diagnosis and treatment method for invasion diagnosis and treatment, especially electrode needle exploration, which often leads to large errors in diagnosis due to deviations in the exploration path. For this reason, the present invention provides an AI-assisted lung cancer pleural invasion diagnosis auxiliary method and system. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention proposes an AI-assisted diagnosis auxiliary method and system for pleural invasion of lung cancer, including: first, obtaining scanning images and electrode needle signals, and obtaining an enhanced training set through preprocessing, time synchronization and enhanced annotation; second, constructing a regional prediction model based on the YOLOv8 algorithm and the attention matching network, and training the model with the enhanced training set and the regional electrical signal conversion error function; third, inputting the real-time scanning image into the trained regional prediction model to obtain a suspected exploration area and an electrode needle exploration prediction signal sequence, and then configuring a safe exploration boundary, combining the real-time electrode needle signal with the prediction signal to obtain the exploration signal error, inputting the configured exploration path correction model for training, and obtaining the optimal exploration path through risk assessment, and finally, integrating the regional prediction model and the exploration path correction model into the electrode needle, and assisting in correcting the electrode needle exploration path based on real-time data, and obtaining the exploration result and invasion exploration score of the suspected exploration area, thereby improving the accuracy and efficiency of the diagnosis of pleural invasion of lung cancer.

[0007] To achieve the above object, the present invention provides the following technical solutions: AI-assisted diagnostic methods for pleural invasion of lung cancer include: S1, obtaining scanning images and electrode needle signals for preprocessing, synchronizing the preprocessed data in time, and enhancing and annotating the scanning images and electrode needle signals to obtain an enhanced training set; S2. Build a regional prediction model based on the YOLOv8 algorithm and the attention matching network, input the enhanced training set into the regional prediction model and use the built-in regional electrical signal conversion error function for training to obtain a trained regional prediction model; S3, inputting the acquired scanned image into the regional prediction model to obtain a suspected detection region and a corresponding electrode needle detection prediction signal sequence; S4, configure a dynamic safety exploration boundary, use the electrode needle signal of the same time dimension as the scan image in S3 and the corresponding electrode needle exploration prediction signal sequence to obtain the exploration signal error, and input the suspected exploration area, safety exploration boundary, and exploration signal error of the scan image into the configured exploration path correction model for training, and obtain simulated multiple exploration paths, and perform exploration risk assessment on the simulated multiple exploration paths to obtain the optimal exploration path; S5. Integrate the trained regional prediction model and exploration path correction model into the electrode needle, and perform auxiliary correction on the electrode needle exploration path according to the real-time acquired scanning image and electrode needle signal to obtain the exploration result of the suspected exploration area and the corresponding invasion exploration score.

[0008] Specifically, the steps for obtaining the enhanced training set include: S101, preprocessing is performed based on the acquired scanned image by using a grayscale algorithm and a filtering denoising algorithm, and a low-frequency baseline drift component in the electrode needle signal is removed by using a high-pass filter to obtain a scanned image and an electrode needle signal after preprocessing; S102, based on the pre-processed scanned image, marking the violated area and the potential violated area by using an expert experience method, and marking the levels of the violated area and the potential violated area, to obtain a scanned image with completed marking; S103, according to the corresponding marked area in the marked scanned image, using an association algorithm to screen and obtain the electrode signal sequence and the corresponding correlation degree corresponding to the same position area from the preprocessed electrode needle signals; S104, pairing the electrode signal sequence corresponding to the same position area with the same position area of ​​the scanned image through the correlation degree to obtain an enhanced training set.

[0009] Specifically, the regional prediction model includes an improved backbone layer, an improved bottleneck layer and a detection layer; the improved backbone layer includes the original first low-dimensional feature extraction sublayer, the second low-dimensional feature extraction sublayer and the improved third to ninth high-dimensional feature extraction sublayers of the MobileViT network, and the improved bottleneck layer includes the original bottleneck sublayer and the ECA channel attention sublayer; the construction and training steps of the regional prediction model include: S201, inputting the enhanced training set into the improved backbone layer, performing feature extraction on the scanned image in the enhanced training set through the first low-dimensional feature extraction sublayer and the second low-dimensional feature extraction sublayer, obtaining the initial image extraction feature, and inputting the electrode signal sequence corresponding to the current input scanned image in the enhanced training set into the one-dimensional convolution sublayer and the dimension adjustment sublayer, obtaining the initial electrode signal feature with the same dimension as the initial image extraction feature; S202, cascade-fusing the initial electrode signal feature and the initial image extraction feature and inputting them into the third to ninth high-dimensional feature extraction sublayers connected in series to obtain a first high-dimensional image-signal extraction feature; S203, inputting the first high-dimensional image-signal extraction feature into the original bottleneck sublayer, performing upsampling feature extraction, and fusing the features extracted from the corresponding sub-blocks in the original bottleneck sublayer with the high-dimensional image-signal extraction features obtained from the corresponding sub-layer in the improved backbone layer through the improved fusion path between the backbone layer and the bottleneck layer in the original YOLOv8 to obtain the first fused image-signal extraction feature; The improved fusion path is achieved by adding a convolution kernel size of 1 in each fusion path between the backbone layer and the bottleneck layer in the original YOLOv8 network. The two-dimensional convolutional network is constructed.

[0010] Specifically, the steps of constructing and training the regional prediction model also include: S204, inputting the first fused image-signal extraction feature into the ECA channel attention sublayer, performing enhanced correlation fusion through the correlation between the electrode signal and the corresponding area in the corresponding scanned image, and obtaining an enhanced fusion feature; S205, inputting the enhanced fusion features into the regional prediction sublayer, regional assessment sublayer and regional electrical signal conversion sublayer in the detection layer respectively, and obtaining the predicted infringed regional anchor frame and anchor frame error, regional infringement assessment level and level error assessment probability and the corresponding predicted anchor frame regional electrode signal and regional electrode signal prediction error in turn; S206, constructing a regional electrical signal conversion error function using the regional electrode signal prediction error, the anchor frame error, and the level error evaluation probability, and using the regional electrical signal conversion error function for training, while obtaining the loss of each round of training; S207, setting a loss threshold and an early stopping value, when the loss of each round of training is less than the loss threshold within a continuous early stopping value range, a trained regional prediction model is obtained.

[0011] Specifically, the steps for constructing the dynamic security exploration boundary include: S401, determining an initial safe exploration boundary interval based on the exploration area, and obtaining structural parameters and respiratory parameters corresponding to each exploration sub-area under the exploration area corresponding to different ages; S402, dividing different ages by expert experience method, and constructing a safe exploration boundary adjustment factor by using the structural parameters and respiratory parameters corresponding to each exploration sub-area under the exploration area corresponding to the divided age groups; S403. Using the safety exploration boundary adjustment factor and the initial safety exploration boundary interval, obtain dynamic safety exploration boundaries corresponding to different age groups.

[0012] Specifically, the training steps of the exploration path correction model include: S411, constructing an exploration path correction model based on the path algorithm, inputting the dynamic safety exploration boundary and the infringing area anchor frame predicted by the scanned image into the exploration path correction model, and generating a dynamic safety exploration boundary. N exploration paths and their corresponding electrode signals and the safety assessment score corresponding to each exploration path; S412, setting a safety assessment score threshold and a path error threshold, and constructing a path error of the exploration area by using the electrode signal corresponding to the exploration path and the electrode signal of the anchor frame area predicted by the regional prediction model; S413. Train the exploration path correction model using the safety assessment score threshold, the path error threshold, the safety assessment score corresponding to each exploration path, and the exploration area path error. When all simulated exploration paths satisfy that the exploration area path error is less than the path error threshold and the safety assessment score corresponding to each exploration path is greater than the safety assessment score threshold, a trained exploration path correction model is obtained.

[0013] Specifically, the steps of obtaining the optimal exploration path include: S414, obtaining a dynamic safety exploration boundary of the exploration object according to the basic information and physiological state information corresponding to the exploration object; S415, inputting the dynamic safety exploration boundary of the exploration object, the predicted suspected exploration area of ​​the exploration object, and the corresponding electrode needle exploration prediction signal sequence into the trained exploration path correction model, and obtaining the exploration path that meets the threshold condition of S413 and has the maximum safety assessment score and the minimum exploration area path error as the optimal exploration path; S416, feeding back the parameters corresponding to the optimal exploration path to the electrode needle, and correcting the real-time exploration path of the electrode needle at the current moment.

[0014] AI-assisted lung cancer pleural invasion diagnosis auxiliary system, including: data processing module, regional prediction module, exploration path correction module and integrated output module; The data processing module includes a data synchronization acquisition unit and a labeling unit; The data synchronization acquisition unit is used to synchronously acquire the scanning image of lung cancer pleural invasion and the electrode needle signal matching the scanning image area; the labeling unit is used to synchronously label the acquired scanning image and electrode needle signal correspondingly through the expert experience method to obtain an enhanced training set; The regional prediction module includes a model training unit and a prediction unit; A model training unit is used to build a regional prediction model based on the YOLOv8 algorithm and the attention matching network, input the enhanced training set into the regional prediction model and use the built-in regional electrical signal conversion error function for training to obtain a trained regional prediction model; The prediction unit is used to input the acquired scanned image into the regional prediction model to obtain a suspected detection area and a corresponding electrode needle detection prediction signal sequence.

[0015] Specifically, the exploration path correction module includes an error acquisition unit, a correction model training unit and an optimal path unit; The error acquisition unit uses the electrode needle signal in the same time dimension as the real-time scanning image and the corresponding electrode needle detection prediction signal sequence to acquire the detection signal error in the same time dimension; A correction model training unit is used to input the suspected detection area, the safe detection boundary, and the detection signal error of the scanned image into the detection path correction model for training, so as to obtain a trained detection path correction model; The optimal path unit is used to correct the model according to the trained exploration path, simulate and generate multiple exploration paths by scanning the suspected exploration area and safe exploration boundary of the image, and perform exploration risk assessment on the simulated multiple exploration paths to obtain the optimal exploration path.

[0016] A computer-readable storage medium stores computer instructions, which, when executed, execute an AI-assisted method for diagnosing pleural invasion of lung cancer.

[0017] Compared with the prior art, the present invention has the following beneficial effects: In view of the deficiencies of the prior art, the present invention constructs a regional prediction model by combining YOLOv8 with an attention matching network, inputs an enhanced training set into the model for training, and optimizes the model performance by using a built-in regional electrical signal conversion error function, which enables the model to more accurately identify suspected exploration areas and reduce misjudgments; secondly, by configuring a safe exploration boundary and combining the real-time electrode needle signal with the predicted signal sequence to obtain the exploration signal error, the accurate correction and risk assessment of the exploration path are achieved, and the problem of diagnostic error caused by path deviation in the prior art is effectively solved, and the optimal exploration path is generated, which greatly improves the safety and accuracy of the operation; finally, the exploration path is dynamically adjusted according to the real-time acquired scanning image and electrode needle signal, and the exploration results of the suspected exploration area and the corresponding invasion exploration score are provided. The real-time feedback mechanism not only helps doctors make more accurate clinical decisions, but also reduces the risks that may occur during the operation. In addition, in the process of generating the exploration path, the present invention also obtains the corresponding dynamic safe exploration boundary through the physiological states of different objects, further ensuring the safety of the exploration object during the exploration process. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of an AI-assisted method for diagnosing pleural invasion of lung cancer in Example 1 of the present invention; Figure 2 This is a structural diagram of a regional prediction model in Example 1 of the present invention; Figure 3 This is a module diagram of an AI-assisted lung cancer pleural invasion diagnosis assistance system according to Example 2 of the present invention. DETAILED DESCRIPTION

[0019] Example 1 See also Figure 1 The present invention provides an embodiment: an AI-assisted diagnosis method for pleural invasion of lung cancer, comprising the following steps: S1, obtaining a scan image and an electrode needle signal for preprocessing, synchronizing the preprocessed data in time, and enhancing and annotating the scan image and the electrode needle signal to obtain an enhanced training set; further, in this embodiment, the scan image is a CT image or an MRI image obtained during the diagnosis of pleural invasion of lung cancer; Furthermore, in this embodiment, the step of acquiring the enhanced training set includes: S101, preprocessing is performed based on the acquired scanned image by using a grayscale algorithm and a filtering denoising algorithm, and a low-frequency baseline drift component in the electrode needle signal is removed by using a high-pass filter to obtain a scanned image and an electrode needle signal after preprocessing; S102, based on the pre-processed scanned image, marking the violated area and the potential violated area by using an expert experience method, and marking the levels of the violated area and the potential violated area, to obtain a scanned image with completed marking; In this embodiment, firstly, the position, size and shape of the lung cancer tumor in the scanned image are marked. This process is achieved by manually outlining the tumor boundary, or using existing medical image segmentation algorithms for automatic or semi-automatic segmentation, and then manual correction. Secondly, the position and shape of the pleura are marked, especially the part of the pleura adjacent to the tumor or potentially invaded. Thirdly, based on medical knowledge and clinical experience, potential areas where lung cancer may invade the pleura are marked, such as the contact interface between the tumor and the pleura, and areas of pleural thickening or nodular changes.

[0020] S103, according to the corresponding marked area in the marked scanned image, using an association algorithm to screen and obtain the electrode signal sequence and the corresponding correlation degree corresponding to the same position area from the preprocessed electrode needle signals; Exemplarily, during the scanning image and electrode needle signal acquisition process, the acquisition timestamp of each data point is recorded, and the scanning image and electrode needle signal in the same time point or time interval are matched by timestamp matching; for example, assuming that the scanning image is acquired at a speed of 10 frames per second and the electrode needle signal is acquired at a speed of 1000 data points per second, the image frame and electrode needle signal data segment corresponding to a certain moment are found according to the timestamp; if the acquisition frequencies of the scanning image and the electrode needle signal are inconsistent, the scanning image and the electrode needle signal are matched by interpolation.

[0021] S104, pairing the electrode signal sequence corresponding to the same position area with the same position area of ​​the scanned image through the correlation degree to obtain an enhanced training set.

[0022] The labels in this embodiment also include information such as whether the tumor has invaded the pleura (yes / no), the degree of invasion (mild, moderate, severe, etc.), and the feature information related to diagnosis in the scan image and electrode needle signal is used as auxiliary labels.

[0023] The process first preprocesses the scanned image with a grayscale algorithm and a filtering denoising algorithm to ensure image clarity and consistency; at the same time, a high-pass filter is used to remove the low-frequency baseline drift component in the electrode needle signal, which effectively reduces noise interference while improving signal purity and enhancing spatiotemporal synchronization; secondly, the expert experience method is used to annotate the invaded areas and potential invaded areas, and their levels are marked to generate detailed annotated scanned images, ensuring the accuracy and authority of the annotations, making the model training closer to the actual clinical situation; thirdly, the association algorithm is used to screen out the electrode signal sequence corresponding to the annotated area and its correlation degree from the preprocessed electrode needle signal. This precise matching of multimodal data not only enriches the information volume of the training samples, but also ensures the consistency and correlation between different modal data, providing more comprehensive data support for model learning; finally, the electrode signal sequence in the same position area is associated and paired with the scanned image through the correlation degree, and an enhanced training set is constructed, achieving a high degree of fusion of multimodal data, and significantly improving the accuracy and generalization ability of the model in identifying lesion areas.

[0024] S2. Build a regional prediction model based on the YOLOv8 algorithm and the attention matching network, input the enhanced training set into the regional prediction model and use the built-in regional electrical signal conversion error function for training to obtain a trained regional prediction model; Furthermore, in this embodiment, the regional prediction model includes an improved backbone layer and an improved bottleneck layer and detection layer; the improved backbone layer includes the original first low-dimensional feature extraction sublayer, the second low-dimensional feature extraction sublayer and the improved third to ninth high-dimensional feature extraction sublayers of the MobileViT network, and the improved bottleneck layer includes the original bottleneck sublayer and the ECA channel attention sublayer; further, in this embodiment, the MobileViT network enhances the network's ability to combine local and global information by integrating the advantages of CNN and ViT.

[0025] Furthermore, in this embodiment, the first low-dimensional feature extraction sublayer and the second low-dimensional feature extraction sublayer are original sublayers in the backbone layer of the original YOLOv8 algorithm, which are constructed through the Darknet-53 network.

[0026] See also Figure 2 Furthermore, in this embodiment, the steps of constructing and training the regional prediction model include: S201, inputting the enhanced training set into the improved backbone layer, performing feature extraction on the scanned image in the enhanced training set through the first low-dimensional feature extraction sublayer and the second low-dimensional feature extraction sublayer, obtaining the initial image extraction feature, and inputting the electrode signal sequence corresponding to the current input scanned image in the enhanced training set into the one-dimensional convolution sublayer and the dimension adjustment sublayer, obtaining the initial electrode signal feature with the same dimension as the initial image extraction feature; S202, cascade-fusing the initial electrode signal feature and the initial image extraction feature and inputting them into the third to ninth high-dimensional feature extraction sublayers connected in series to obtain a first high-dimensional image-signal extraction feature; S203, inputting the first high-dimensional image-signal extraction feature into the original bottleneck sublayer, performing upsampling feature extraction, and fusing the features extracted from the corresponding sub-blocks in the original bottleneck sublayer with the high-dimensional image-signal extraction features obtained from the corresponding sub-layer in the improved backbone layer through the improved fusion path between the backbone layer and the bottleneck layer in the original YOLOv8 to obtain the first fused image-signal extraction feature; Furthermore, in this embodiment, the improved fusion path is achieved by adding a convolution kernel size of 100 in each fusion path between the backbone layer and the bottleneck layer in the original YOLOv8 network. The two-dimensional convolutional network is constructed; the improved fusion path includes the first fusion path, the second fusion path and the third fusion path, and the corresponding structure is the same as the fusion path structure between the backbone layer and the bottleneck layer in the original YOLOv8 network.

[0027] S204, inputting the first fused image-signal extraction feature into the ECA channel attention sublayer, performing enhanced correlation fusion through the correlation between the electrode signal and the corresponding area in the corresponding scanned image, and obtaining an enhanced fusion feature; S205, inputting the enhanced fusion features into the regional prediction sublayer, regional assessment sublayer and regional electrical signal conversion sublayer in the detection layer respectively, and obtaining the predicted infringed regional anchor frame and anchor frame error, regional infringement assessment level and level error assessment probability and the corresponding predicted anchor frame regional electrode signal and regional electrode signal prediction error in turn; Furthermore, in this embodiment, the regional prediction sublayer is constructed by the YOLOv8 original prediction head; the regional evaluation sublayer is constructed by the comprehensive fuzzy evaluation algorithm; the regional electrical signal conversion sublayer is constructed by the fully connected network; S206, constructing a regional electrical signal conversion error function using the regional electrode signal prediction error, the anchor frame error, and the level error evaluation probability, and using the regional electrical signal conversion error function for training, while obtaining the loss of each round of training; S207, setting a loss threshold and an early stop value, when the loss of each round of training is less than the loss threshold within the continuous early stop value range, a trained regional prediction model is obtained. Further, in this embodiment, the early stop value is set to 10.

[0028] In the training process, the improved backbone layer performs preliminary feature extraction on the scanned image through the first low-dimensional feature extraction sublayer and the second low-dimensional feature extraction sublayer, and combines the one-dimensional convolution sublayer and the dimension adjustment sublayer to process the electrode signal sequence to obtain the initial electrode signal features with the same dimension as the scanned image features, ensuring the effective fusion of multimodal data in the early stage; secondly, the initial electrode signal features and the initial image extraction features are cascaded and input into the high-dimensional feature extraction sublayer, and the ECA channel attention mechanism is introduced to optimize the bottleneck layer, realizing the efficient extraction and fusion of deep-level features. The improved fusion path is improved by adding a convolution kernel size of The two-dimensional convolutional network not only enhances the correlation between features, but also reduces the number of model parameters, realizes cross-channel information integration and increases nonlinear characteristics, effectively integrates multi-scale information, and further improves the model's ability to identify complex lesion areas; thirdly, the ECA channel attention sublayer enhances the correlation fusion, so that the model can more accurately capture the relationship between the electrode signal and the corresponding area in the scanned image, generate enhanced fusion features, and the ECA channel attention mechanism of the bottleneck layer also enhances the capture of long-distance dependencies, improving the feature expression and fusion capabilities of the network; then, the regional prediction sublayer, regional evaluation sublayer and regional electrical signal conversion sublayer in the detection layer are responsible for the prediction of the anchor box of the infringing area, the prediction of the evaluation level and the prediction of the electrode signal, respectively, providing comprehensive diagnostic information; finally, the regional electrical signal conversion error function is constructed using the regional electrical signal prediction error, anchor box error and level error evaluation probability, and the training is performed by setting the loss threshold and early stopping value, which not only ensures that the model converges to the optimal solution, but also improves the generalization ability of the model, and effectively avoids the overfitting problem, ensuring the stability and reliability of the model in practical applications.

[0029] S3, inputting the acquired scanned image into the regional prediction model to obtain a suspected detection region and a corresponding electrode needle detection prediction signal sequence; S4, configure a dynamic safety exploration boundary, use the electrode needle signal of the same time dimension as the scan image in S3 and the corresponding electrode needle exploration prediction signal sequence to obtain the exploration signal error, and input the suspected exploration area, safety exploration boundary, and exploration signal error of the scan image into the configured exploration path correction model for training, and obtain simulated multiple exploration paths, and perform exploration risk assessment on the simulated multiple exploration paths to obtain the optimal exploration path; Furthermore, in this embodiment, the steps of constructing the dynamic security exploration boundary include: S401, determining an initial safe exploration boundary interval based on the exploration area, and obtaining structural parameters and respiratory parameters corresponding to each exploration sub-area under the exploration area corresponding to different ages; Furthermore, the initial safety exploration boundary interval in this embodiment is obtained by combining the safety exploration boundary and exploration area of ​​historical cases through an expert experience algorithm.

[0030] Furthermore, different exploration areas (such as the area of ​​the lung used for the diagnosis of pleural invasion of lung cancer) have specific structural and functional characteristics. For example, the lungs have specific blood vessels, trachea, and nerve distributions. The location and function of these structures determine that the electrode needle cannot enter certain areas at will during exploration, otherwise it may cause serious damage. By understanding this knowledge, a preliminary and generally applicable safe exploration range can be determined, that is, the initial safe exploration boundary interval. At the same time, the physiological state of the human body at different ages is different. The structural parameters (such as tissue elasticity, thickness, etc.) and respiratory parameters (such as respiratory frequency, amplitude, etc.) corresponding to each exploration sub-area under the corresponding exploration area of ​​different ages are obtained. These parameters reflect the physiological differences of the human body at different ages in the exploration area, providing data support for the subsequent construction of regulatory factors.

[0031] S402, dividing different ages by expert experience method, and constructing a safe exploration boundary adjustment factor by using the structural parameters and respiratory parameters corresponding to each exploration sub-area under the exploration area corresponding to the divided age groups; Furthermore, there are significant differences in physiological functions, tissue structures, etc. between people of different age groups, and these differences will affect the setting of the safe exploration boundary. By using the structural parameters and respiratory parameters of each exploration sub-area under the exploration area corresponding to the divided age groups, a safe exploration boundary adjustment factor is constructed through a certain mathematical model or empirical formula; this adjustment factor can quantify the impact of physiological differences in different age groups on the safe exploration boundary. For example, aging may lead to a decrease in tissue elasticity and changes in respiratory function, and the adjustment factor will make corresponding adjustments to the safe exploration boundary based on these changes.

[0032] S403. Using the safety exploration boundary adjustment factor and the initial safety exploration boundary interval, obtain dynamic safety exploration boundaries corresponding to different age groups.

[0033] Furthermore, the constructed safe exploration boundary adjustment factor is applied to the initial safe exploration boundary interval to obtain the dynamic safe exploration boundary corresponding to different age groups; this dynamic adjustment takes into account the physiological characteristics of different age groups, making the safe exploration boundary more in line with the actual individual situation and improving the safety and accuracy of the exploration.

[0034] Exemplarily, based on knowledge of lung anatomy, the locations of important structures such as the large blood vessels and trachea of ​​the lungs are known. To prevent the electrode needle from damaging these structures, the initial safe exploration boundary interval is set to an area at least 5 mm away from the large blood vessels and trachea of ​​the lungs. At the same time, the structural parameters of each exploration sub-area corresponding to this exploration area are obtained for different ages (such as children, adults, and the elderly) (such as children's lung tissue has better elasticity, with an elasticity coefficient of 0.8; the elderly's lung tissue has poor elasticity, with an elasticity coefficient of 0.4) and respiratory parameters (such as children's faster breathing rate, 25 times per minute; adults' 15 times per minute; and the elderly's 18 times per minute).

[0035] Medical experts divide age groups into children (0-12 years old), adults (13-60 years old), and the elderly (61 years old and above) based on experience. For children, because their lung tissue is relatively tender and their respiratory rate is fast, the safe exploration boundary adjustment factor is constructed as 1.2 based on the structural parameters and respiratory parameters of their exploration sub-areas (which means that the safety range is appropriately expanded based on the initial boundary). For the elderly, taking into account factors such as decreased elasticity of lung tissue and weakened respiratory function, the adjustment factor is constructed as 0.8 (which means that the safety range is appropriately narrowed, because the elderly's lungs have poor tolerance to damage and need to be operated with more caution).

[0036] For children, the initial safe exploration boundary interval is an area at least 5 mm away from the large pulmonary vessels and trachea. After applying the adjustment factor 1.2, the dynamic safe exploration boundary becomes an area at least 6 mm (5×1.2 = 6) away from the large pulmonary vessels and trachea. For the elderly, after applying the adjustment factor 0.8, the dynamic safe exploration boundary becomes an area at least 4 mm (5×0.8 = 4) away from the large pulmonary vessels and trachea.

[0037] Furthermore, in this embodiment, the training step of the exploration path correction model includes: S411, constructing an exploration path correction model based on the path algorithm, inputting the dynamic safety exploration boundary and the infringing area anchor frame predicted by the scanned image into the exploration path correction model, and generating a dynamic safety exploration boundary. N exploration paths and their corresponding electrode signals and the safety assessment score corresponding to each exploration path; S412, setting a safety assessment score threshold and a path error threshold, and constructing a path error of the exploration area by using the electrode signal corresponding to the exploration path and the electrode signal of the anchor frame area predicted by the regional prediction model; S413. Train the exploration path correction model using the safety assessment score threshold, the path error threshold, the safety assessment score corresponding to each exploration path, and the exploration area path error. When all simulated exploration paths satisfy that the exploration area path error is less than the path error threshold and the safety assessment score corresponding to each exploration path is greater than the safety assessment score threshold, a trained exploration path correction model is obtained.

[0038] Furthermore, in this embodiment, the step of obtaining the optimal exploration path includes: S414, obtaining a dynamic safety exploration boundary of the exploration object according to the basic information and physiological state information corresponding to the exploration object; S415, inputting the dynamic safety exploration boundary of the exploration object, the predicted suspected exploration area of ​​the exploration object, and the corresponding electrode needle exploration prediction signal sequence into the trained exploration path correction model, and obtaining the exploration path that meets the threshold condition of S413 and has the maximum safety assessment score and the minimum exploration area path error as the optimal exploration path; S416, feeding back the parameters corresponding to the optimal exploration path to the electrode needle, and correcting the real-time exploration path of the electrode needle at the current moment.

[0039] This process determines the initial safe exploration boundary interval based on anatomical and physiological knowledge, and obtains the structural parameters and respiratory parameters corresponding to different age groups, ensuring the basic scientific nature of the exploration boundary setting; secondly, the expert experience method is used to divide different age groups, and the safe exploration boundary adjustment factor is constructed according to the structural parameters and respiratory parameters of each age group. This process quantifies the impact of physiological differences in different age groups on the exploration boundary, making the safety boundary more in line with the actual situation of the individual. For example, children have tender tissues and fast breathing rates, so their adjustment factor is set to 1.2, and the safety range is appropriately expanded; the elderly have decreased tissue elasticity and weakened respiratory function, so the adjustment factor is set to 0.8, narrowing the safety range to improve the caution of operation; thirdly, the adjustment factor is applied to the initial safe exploration boundary interval to obtain a dynamic safe exploration boundary. This dynamic adjustment takes into account individual physiological characteristics and improves the safety and accuracy of exploration. In the example, the dynamic safe exploration boundary of children is expanded from 5mm to 6mm, while that of the elderly is reduced from 5mm to 4mm, reflecting personalized safety considerations.

[0040] During the training process of the exploration path correction model, the exploration path can be optimized through path algorithm, threshold setting and error evaluation to ensure the performance of the training model and meet clinical safety and accuracy requirements. The optimal exploration path is obtained through the exploration path correction model and fed back to correct the real-time exploration path of the electrode needle, which can ensure that the exploration operation is more accurate and safe, and realize system optimization from data processing, model training to actual operation, which helps to improve the reliability and effectiveness of diagnosis and treatment. In addition, the systematic process and parameter adjustment mechanism of this process fully consider individual differences, and can provide more targeted exploration services for different patients, thereby improving the quality and effect of the entire diagnostic process.

[0041] S5. Integrate the trained regional prediction model and exploration path correction model into the electrode needle, and perform auxiliary correction on the electrode needle exploration path according to the real-time acquired scanning image and electrode needle signal to obtain the exploration result of the suspected exploration area and the corresponding invasion exploration score.

[0042] Furthermore, in this embodiment, the detection result includes whether it is an infringement area, and an infringement detection score is obtained through an evaluation algorithm according to specific parameters of the corresponding infringement area.

[0043] Example 2 See also Figure 3 , another embodiment provided by the present invention: an AI-assisted lung cancer pleural invasion diagnosis auxiliary system, comprising: a data processing module, a region prediction module, a detection path correction module and an integrated output module; Data processing module, used for acquisition and enhancement preprocessing of lung cancer pleural invasion scanning images; The data processing module includes a data synchronization acquisition unit and a labeling unit; A data synchronous acquisition unit is used to synchronously acquire a scan image of lung cancer pleural invasion and an electrode needle signal matching the scan image area; a labeling unit is used to synchronously label the acquired scan image and the electrode needle signal by an expert experience method to obtain an enhanced training set; further, in this embodiment, the area corresponding to the scan image and the electrode signal obtained by exploring the same area in the electrode needle signal are paired accordingly; The regional prediction module is used for model training and obtaining the suspected detection area and the corresponding electrode needle detection prediction signal sequence; The regional prediction module includes a model training unit and a prediction unit; A model training unit is used to build a regional prediction model based on the YOLOv8 algorithm and the attention matching network, input the enhanced training set into the regional prediction model and use the built-in regional electrical signal conversion error function for training to obtain a trained regional prediction model; A prediction unit, used to input the acquired scanned image into a regional prediction model to obtain a suspected detection region and a corresponding electrode needle detection prediction signal sequence; An exploration path correction module is used to train the exploration path correction model and obtain the optimal exploration path; The exploration path correction module includes an error acquisition unit, a correction model training unit and an optimal path unit; The error acquisition unit uses the electrode needle signal in the same time dimension as the real-time scanning image and the corresponding electrode needle detection prediction signal sequence to acquire the detection signal error in the same time dimension; A correction model training unit is used to input the suspected detection area, the safe detection boundary, and the detection signal error of the scanned image into the detection path correction model for training, so as to obtain a trained detection path correction model; The optimal path unit is used to modify the model according to the trained exploration path, simulate and generate multiple exploration paths by scanning the suspected exploration area and the safe exploration boundary of the image, and perform exploration risk assessment on the simulated multiple exploration paths to obtain the optimal exploration path; The integrated output module is used to integrate the trained regional prediction model and exploration path correction model into the electrode needle, and to assist in correcting the electrode needle exploration path according to the real-time acquired scanning image and electrode needle signal to obtain the exploration results of the suspected exploration area and the corresponding invasion exploration score.

[0044] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present invention and the claims, and all of these are within the protection of the present invention.

[0045] If the disclosed technical solution involves personal information, the product using the disclosed technical solution has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the disclosed technical solution involves sensitive personal information, the product using the disclosed technical solution has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, when the personal information processing rules are notified by obvious signs / information, the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

Claims

1. An AI-assisted method for diagnosing pleural invasion of lung cancer, characterized in that: include: S1, obtaining scanning images and electrode needle signals for preprocessing, synchronizing the preprocessed data in time, and enhancing and annotating the scanning images and electrode needle signals to obtain an enhanced training set; S2. Build a regional prediction model based on the YOLOv8 algorithm and the attention matching network, input the enhanced training set into the regional prediction model and use the built-in regional electrical signal conversion error function for training to obtain a trained regional prediction model; S3, inputting the acquired scanned image into the regional prediction model to obtain a suspected detection region and a corresponding electrode needle detection prediction signal sequence; S4, configure a dynamic safety exploration boundary, use the electrode needle signal of the same time dimension as the scan image in S3 and the corresponding electrode needle exploration prediction signal sequence to obtain the exploration signal error, and input the suspected exploration area, safety exploration boundary, and exploration signal error of the scan image into the configured exploration path correction model for training, and obtain simulated multiple exploration paths, and perform exploration risk assessment on the simulated multiple exploration paths to obtain the optimal exploration path; S5. Integrate the trained regional prediction model and exploration path correction model into the electrode needle, and perform auxiliary correction on the electrode needle exploration path according to the real-time acquired scanning image and electrode needle signal to obtain the exploration result of the suspected exploration area and the corresponding invasion exploration score.

2. The AI-assisted diagnosis method for pleural invasion of lung cancer according to claim 1, characterized in that: The step of acquiring the enhanced training set includes: S101, preprocessing is performed based on the acquired scanned image by using a grayscale algorithm and a filtering denoising algorithm, and a low-frequency baseline drift component in the electrode needle signal is removed by using a high-pass filter to obtain a scanned image and an electrode needle signal after preprocessing; S102, based on the pre-processed scanned image, marking the violated area and the potential violated area by using an expert experience method, and marking the levels of the violated area and the potential violated area, to obtain a scanned image with completed marking; S103, according to the corresponding marked area in the marked scanned image, using an association algorithm to screen and obtain the electrode signal sequence and the corresponding correlation degree corresponding to the same position area from the preprocessed electrode needle signals; S104, pairing the electrode signal sequence corresponding to the same position area with the same position area of ​​the scanned image through the correlation degree to obtain an enhanced training set.

3. The AI-assisted diagnosis method for pleural invasion of lung cancer according to claim 2, characterized in that: The regional prediction model includes an improved backbone layer, an improved bottleneck layer and a detection layer; the improved backbone layer includes an original first low-dimensional feature extraction sublayer, a second low-dimensional feature extraction sublayer and the third to ninth high-dimensional feature extraction sublayers improved by the MobileViT network, and the improved bottleneck layer includes an original bottleneck sublayer and an ECA channel attention sublayer; The steps of constructing and training the regional prediction model include: S201, inputting the enhanced training set into the improved backbone layer, performing feature extraction on the scanned image in the enhanced training set through the first low-dimensional feature extraction sublayer and the second low-dimensional feature extraction sublayer, obtaining initial image extraction features, and inputting the electrode signal sequence corresponding to the current input scanned image in the enhanced training set into the one-dimensional convolution sublayer and the dimension adjustment sublayer, obtaining initial electrode signal features of the same dimension as the initial image extraction features; S202, cascade-fusing the initial electrode signal feature and the initial image extraction feature and inputting them into the third to ninth high-dimensional feature extraction sublayers connected in series to obtain a first high-dimensional image-signal extraction feature; S203, inputting the first high-dimensional image-signal extraction feature into the original bottleneck sublayer, performing upsampling feature extraction, and fusing the features extracted from the corresponding sub-blocks in the original bottleneck sublayer with the high-dimensional image-signal extraction features obtained from the corresponding sub-layer in the improved backbone layer through the improved fusion path between the backbone layer and the bottleneck layer in the original YOLOv8 to obtain the first fused image-signal extraction feature; The improved fusion path is achieved by adding a convolution kernel size of The two-dimensional convolutional network is constructed.

4. The AI-assisted diagnosis method for pleural invasion of lung cancer according to claim 3, characterized in that: The steps of constructing and training the regional prediction model also include: S204, inputting the first fused image-signal extraction feature into the ECA channel attention sublayer, performing enhanced correlation fusion through the correlation between the electrode signal and the corresponding area in the corresponding scanned image, and obtaining an enhanced fusion feature; S205, inputting the enhanced fusion features into the regional prediction sublayer, the regional assessment sublayer and the regional electrical signal conversion sublayer in the detection layer respectively, and obtaining the predicted infringed region anchor frame and anchor frame error, the regional infringed assessment level and the level error assessment probability, and the corresponding predicted anchor frame region electrode signal and the regional electrode signal prediction error in turn; S206, constructing a regional electrical signal conversion error function using the regional electrode signal prediction error, the anchor frame error, and the level error evaluation probability, and using the regional electrical signal conversion error function for training, while obtaining the loss of each round of training; S207, setting a loss threshold and an early stopping value, when the loss of each round of training is less than the loss threshold within a continuous early stopping value range, a trained regional prediction model is obtained.

5. The AI-assisted diagnosis method for pleural invasion of lung cancer according to claim 4, characterized in that: The steps of constructing the dynamic security exploration boundary include: S401, determining an initial safe exploration boundary interval based on the exploration area, and obtaining structural parameters and respiratory parameters corresponding to each exploration sub-area under the exploration area corresponding to different ages; S402, dividing different ages by expert experience method, and constructing a safe exploration boundary adjustment factor by using the structural parameters and respiratory parameters corresponding to each exploration sub-area under the exploration area corresponding to the divided age groups; S403. Using the safety exploration boundary adjustment factor and the initial safety exploration boundary interval, obtain dynamic safety exploration boundaries corresponding to different age groups.

6. The AI-assisted diagnosis auxiliary method for lung cancer pleural invasion according to claim 5, characterized in that: The training steps of the exploration path correction model include: S411, constructing an exploration path correction model based on the path algorithm, inputting the dynamic safety exploration boundary and the infringing area anchor frame predicted by the scanned image into the exploration path correction model, and generating a dynamic safety exploration boundary. N exploration paths and their corresponding electrode signals and the safety assessment score corresponding to each exploration path; S412, setting a safety assessment score threshold and a path error threshold, and constructing a path error of the exploration area by using the electrode signal corresponding to the exploration path and the electrode signal of the anchor frame area predicted by the regional prediction model; S413. Train the exploration path correction model using the safety assessment score threshold, the path error threshold, the safety assessment score corresponding to each exploration path, and the exploration area path error. When all simulated exploration paths satisfy that the exploration area path error is less than the path error threshold and the safety assessment score corresponding to each exploration path is greater than the safety assessment score threshold, a trained exploration path correction model is obtained.

7. The AI-assisted diagnosis method for pleural invasion of lung cancer according to claim 6, characterized in that: The step of obtaining the optimal exploration path includes: S414, obtaining a dynamic safety exploration boundary of the exploration object according to the basic information and physiological state information corresponding to the exploration object; S415, inputting the dynamic safety exploration boundary of the exploration object, the predicted suspected exploration area of ​​the exploration object, and the corresponding electrode needle exploration prediction signal sequence into the trained exploration path correction model, and obtaining the exploration path that meets the threshold condition of S413 and has the maximum safety assessment score and the minimum exploration area path error as the optimal exploration path; S416, feeding back the parameters corresponding to the optimal exploration path to the electrode needle, and correcting the real-time exploration path of the electrode needle at the current moment.

8. An AI-assisted lung cancer pleural invasion diagnosis auxiliary system, which is used to implement the AI-assisted lung cancer pleural invasion diagnosis auxiliary method according to any one of claims 1 to 7, characterized in that: include: Data processing module, area prediction module, exploration path correction module and integrated output module; The data processing module includes a data synchronization acquisition unit and a labeling unit; The data synchronous acquisition unit is used to synchronously acquire the scanned image of lung cancer pleural invasion and the electrode needle signal matching the scanned image area; the labeling unit is used to perform corresponding synchronous labeling on the acquired scanned image and the electrode needle signal through the expert experience method to obtain an enhanced training set; The regional prediction module includes a model training unit and a prediction unit; The model training unit is used to build a regional prediction model based on the YOLOv8 algorithm and the attention matching network, input the enhanced training set into the regional prediction model and use the built-in regional electrical signal conversion error function for training to obtain a trained regional prediction model; The prediction unit is used to input the acquired scanned image into the regional prediction model to obtain a suspected detection region and a corresponding electrode needle detection prediction signal sequence.

9. The AI-assisted lung cancer pleural invasion diagnosis auxiliary system according to claim 8, characterized in that: The exploration path correction module includes an error acquisition unit, a correction model training unit and an optimal path unit; The error acquisition unit uses the electrode needle signal in the same time dimension as the real-time scanning image and the corresponding electrode needle detection prediction signal sequence to acquire the detection signal error in the same time dimension; The correction model training unit is used to input the suspected detection area, the safe detection boundary, and the detection signal error of the scanned image into the detection path correction model for training, so as to obtain a trained detection path correction model; The optimal path unit is used to correct the model of the trained exploration path, simulate and generate multiple exploration paths by scanning the suspected exploration area and the safe exploration boundary of the image, and perform exploration risk assessment on the simulated multiple exploration paths to obtain the optimal exploration path.

10. A computer-readable storage medium, characterized in that: Computer instructions are stored thereon, and when the computer instructions are executed, the AI-assisted lung cancer pleural invasion diagnosis auxiliary method according to any one of claims 1 to 7 is executed.

Images