Hardener intervention infusion dose determination method and system

Abstract

The embodiment of the invention provides a hardening agent intervention infusion dose determination method and system, and belongs to the technical field of artificial intelligence. The method comprises the steps of obtaining a target CT image corresponding to a target organ of a target patient, and performing anomaly recognition on the target CT image by using an anomaly recognition model to obtain a target anomaly region corresponding to the target CT image; determining an associated organ corresponding to the target organ, and calculating a first association distance between the target abnormal region and a target region corresponding to the associated organ; performing size measurement on the target abnormal region to obtain a first target volume of the target abnormal region; and according to the first target volume and the first association distance, determining a corresponding hardener dosage when the target patient performs hardener intervention infusion on the target abnormal area. The problem that in the related technology, when the hardening agent interventional infusion dosage is determined, although the CT-guided image technology provides important guidance for hardening agent interventional infusion, significant challenges still exist in the aspect of accurate determination of the hardening agent dosage is solved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and system for determining a sclerosant interventional infusion dose. Background Art

[0002] Thoracic malignancies such as breast cancer, esophageal cancer, and lung cancer all have proximal metastasis to varying degrees. In addition to chemotherapy, radiotherapy, targeted therapy, and immunotherapy, for tumors with clear boundaries and internal vascular structures (sometimes contrast agents are needed to assist in judgment), CT-guided percutaneous intratumoral sclerosing agent injection therapy can sometimes be used. Through short-term injection of sufficient doses, proteins are denatured and coagulative necrosis of tumor tissue is induced. The vascular epithelial cells around the tumor can also necrotize and form blood clots, further promoting overall ischemic necrosis of the tumor tissue.

[0003] Currently, the commonly used sclerosants are mainly povidone-iodine and anhydrous ethanol, among which percutaneous anhydrous ethanol injection therapy (PEI) is the most intensively used. Ethanol also has some side effects during interventional infusion therapy, severe of which may cause tissue damage or even life-threatening. According to CT-guided images (including both tumor tissue shadows and vascular shadows), the reasonable determination of the sclerosant interventional infusion dose can effectively reduce the probability of complications including tissue necrosis, peripheral nerve injury, acute alcohol poisoning, hemolysis of red blood cells, pulmonary embolism, right heart failure, arrhythmia and even death, and protect the health and safety of patients. However, in the related technology, although CT-guided imaging technology provides important guidance for sclerosant interventional infusion, there are still significant challenges in accurately determining the sclerosant dose.

[0004] Therefore, it is urgent to design a technical solution to solve at least one of the above technical problems. Summary of the invention

[0005] The main purpose of the embodiments of the present invention is to provide a method and system for determining the dosage of sclerosant interventional infusion, aiming to solve the problem in the related art that although CT-guided imaging technology provides important guidance for sclerosant interventional infusion, there are still significant challenges in accurately determining the dosage of sclerosant.

[0006] In a first aspect, an embodiment of the present invention provides a method for determining a dosage of a sclerosant interventional infusion, comprising:

[0007] Obtaining a target CT image corresponding to a target organ of a target patient, and performing abnormality recognition on the target CT image using an abnormality recognition model to obtain a target abnormal region corresponding to the target CT image;

[0008] Determining an associated organ corresponding to the target organ, and calculating a first associated distance between the target abnormal region and the target region corresponding to the associated organ;

[0009] Measuring the size of the target abnormal area to obtain a first target volume corresponding to the target abnormal area;

[0010] The corresponding sclerosant dosage when the target patient performs sclerosant interventional infusion on the target abnormal area is determined according to the first target volume and the first associated distance.

[0011] In a second aspect, an embodiment of the present invention provides a system for determining a dosage of a sclerosant interventional infusion, comprising:

[0012] An abnormality recognition module is used to obtain a target CT image corresponding to a target organ of a target patient, and use an abnormality recognition model to perform abnormality recognition on the target CT image to obtain a target abnormal area corresponding to the target CT image;

[0013] a distance calculation module, used to determine an associated organ corresponding to the target organ, and calculate a first associated distance between the target abnormal region and the target region corresponding to the associated organ;

[0014] A volume calculation module, used for measuring the size of the target abnormal area to obtain a first target volume corresponding to the target abnormal area;

[0015] The dosage determination module is used to determine the corresponding dosage of the sclerosant when the target patient performs sclerosant interventional infusion on the target abnormal area according to the first target volume and the first associated distance.

[0016] In a third aspect, an embodiment of the present invention further provides a terminal device, comprising a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the steps of any method for determining the interventional infusion dose of a sclerosant as provided in the specification of the present invention are realized.

[0017] In a fourth aspect, an embodiment of the present invention further provides a storage medium for computer-readable storage, characterized in that the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any method for determining the interventional infusion dose of a sclerosant as provided in the specification of the present invention.

[0018] The embodiment of the present invention provides a method and system for determining the dose of sclerosant interventional infusion, the method comprising: obtaining a target CT image corresponding to a target organ of a target patient, and then using an abnormality recognition model to deeply analyze the target CT image to obtain a target abnormal area. Further, by determining the associated organ of the target organ and calculating the first associated distance between the target abnormal area and the target area corresponding to the associated organ, the spatial relationship between the target abnormal area and the target area is quantified and understood, so as to provide support for the subsequent accuracy and safety of the dose and minimize the risk to the target patient, and then the size of the target abnormal area is measured, and its first target volume can be accurately calculated. Finally, the first target volume and the first associated distance are combined for analysis to form a more comprehensive and three-dimensional treatment decision model to determine the corresponding sclerosant dose when the target patient performs sclerosant interventional infusion on the target abnormal area, so that the effective treatment of the target abnormal area can be ensured through a reasonable sclerosant dose, while avoiding potential side effects caused by excessive or insufficient doses. The problem that in the related art, although CT-guided imaging technology provides important guidance for sclerosant interventional infusion, there are still significant challenges in accurately determining the sclerosant dose is solved. The reliability and accuracy of the prediction of the corresponding sclerosant dose when the sclerosant is injected into the target abnormal area is improved, while ensuring the treatment effect and improving the safety of the target patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A schematic flow chart of a method for determining a dosage of a sclerosant interventional infusion provided by an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of the module structure of a system for determining the dosage of a sclerosant interventional infusion provided by an embodiment of the present invention;

[0022] Figure 3 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0025] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0026] The embodiment of the present invention provides a method and system for determining the dosage of a sclerosant interventional infusion. The method for determining the dosage of a sclerosant interventional infusion can be applied to a terminal device, which can be an electronic device such as a tablet computer, a laptop computer, a desktop computer, a personal digital assistant, and a wearable device. The terminal device can be a server or a server cluster.

[0027] Thoracic malignancies such as breast cancer, esophageal cancer, and lung cancer all have proximal metastasis to varying degrees. In addition to chemotherapy, radiotherapy, targeted therapy, and immunotherapy, for tumors with clear boundaries and internal vascular structures (sometimes contrast agents are needed to assist in judgment), CT-guided percutaneous intratumoral sclerosing agent injection therapy can sometimes be used. Through short-term injection of sufficient doses, proteins are denatured and coagulative necrosis of tumor tissue is induced. The vascular epithelial cells around the tumor can also necrotize and form blood clots, further promoting overall ischemic necrosis of the tumor tissue.

[0028] Currently, the commonly used sclerosants are mainly povidone-iodine and anhydrous ethanol, among which percutaneous anhydrous ethanol injection therapy (PEI) is the most intensively used. Ethanol also has some side effects during interventional infusion therapy, severe of which may cause tissue damage or even life-threatening. According to CT-guided images (including both tumor tissue shadows and vascular shadows), the reasonable determination of the sclerosant interventional infusion dose can effectively reduce the probability of complications including tissue necrosis, peripheral nerve injury, acute alcohol poisoning, hemolysis of red blood cells, pulmonary embolism, right heart failure, arrhythmia and even death, and protect the health and safety of patients. However, in the related technology, although CT-guided imaging technology provides important guidance for sclerosant interventional infusion, there are still significant challenges in accurately determining the sclerosant dose.

[0029] To this end, an embodiment of the present invention provides a method and system for determining the dose of sclerosant interventional infusion, the method comprising: obtaining a target CT image corresponding to the target organ of the target patient, and then using an abnormality recognition model to deeply analyze the target CT image to obtain a target abnormal area. Further, by determining the associated organ of the target organ and calculating the first associated distance between the target abnormal area and the target area corresponding to the associated organ, the spatial relationship between the target abnormal area and the target area is quantified and understood, so as to provide support for the subsequent accuracy and safety of the dose to minimize the risk to the target patient, and then the size of the target abnormal area is measured, and its first target volume can be accurately calculated. Finally, the first target volume and the first associated distance are combined for analysis to form a more comprehensive and three-dimensional treatment decision model to determine the corresponding sclerosant dose when the target patient performs sclerosant interventional infusion on the target abnormal area, so that the effective treatment of the target abnormal area can be ensured through a reasonable sclerosant dose, while avoiding potential side effects caused by excessive or insufficient doses. The problem that although CT-guided imaging technology provides important guidance for sclerosant interventional infusion in the related art, there are still significant challenges in accurately determining the sclerosant dose is solved. The reliability and accuracy of the prediction of the corresponding sclerosant dose when the sclerosant is injected into the target abnormal area is improved, while ensuring the treatment effect and improving the safety of the target patient.

[0030] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0031] Please refer to Figure 1 , Figure 1A schematic flow chart of a method for determining a sclerosant interventional infusion dose provided in an embodiment of the present invention.

[0032] like Figure 1 As shown, the method for determining the sclerosant intervention infusion dose includes steps S101 to S104.

[0033] Step S101: obtaining a target CT image corresponding to a target organ of a target patient, and performing abnormality recognition on the target CT image using an abnormality recognition model to obtain a target abnormal region corresponding to the target CT image.

[0034] For example, target patients who need CT scans are selected through an electronic health record system or a clinical information system. The target organs that need attention are determined according to the specific condition of the target patients, so that a reasonable CT scan plan is formulated according to the characteristics of the target organs, including scanning parameters such as voltage, dose, slice thickness, etc., and then a CT scan is performed in a radiology department or imaging center to obtain the corresponding target CT image.

[0035] Exemplarily, the abnormality recognition model can be a model based on machine learning or deep learning, such as a convolutional neural network model. Thus, the target CT image is input into the abnormality recognition model and the model is run for reasoning to identify target abnormality information that may be abnormal on the target CT image. The target abnormality information includes the location, size and category of the abnormality. Thus, the target abnormal area corresponding to the target CT image is determined according to the target abnormality information.

[0036] In some embodiments, the use of an abnormality recognition model to perform abnormality recognition on the target CT image to obtain a target abnormal region corresponding to the target CT image includes: using an encoder of the abnormality recognition model to extract features of the target CT image step by step to obtain a first encoding feature; using an attention layer of the abnormality recognition model to extract pixel position-related features of the target CT image to obtain a second encoding feature; using a decoder of the abnormality recognition model to perform feature fusion and feature step-by-step restoration using the first encoding feature and the second encoding feature to obtain a target image feature; and performing pixel classification on the target image features according to a pixel classification layer of the abnormality recognition model to obtain the target abnormal region corresponding to the target CT image.

[0037] Exemplarily, the anomaly recognition model includes an encoder, an attention layer, a decoder, and a pixel classification layer.

[0038] Exemplarily, the role of the encoder is to extract high-level features from the target CT image layer by layer. The encoder processes the target CT image through multiple layers of convolution or other neural network layers. Each layer extracts features in a smaller dimension, gradually captures information at different levels of the target CT image, and records feature maps at each layer of the encoder, which is the extraction result of the target CT image features at each layer. Finally, the last layer output of the encoder is determined as the first encoded feature. The first encoded feature represents the most critical, high-level features learned by the target CT image during the feature extraction process.

[0039] Exemplarily, the preprocessed target CT image is input into the attention layer and the attention mechanism is used to dynamically calculate weighted features according to the position importance of each pixel in the target CT image. These weighted features can better reflect the relationship between each pixel in the target CT image, and the output features of the attention layer are merged to obtain the second coded features. The second coded features contain important information about the pixel position in the target CT image and reflect the contextual relationship between local information and global information in the target CT image.

[0040] Exemplarily, the first encoding feature and the second encoding feature are input into the decoder, and the features are gradually restored to the target image features at the same image size as the target CT image through linear interpolation and feature fusion operations.

[0041] Exemplarily, the pixel classification layer is an activation function, and the activation function is used to perform abnormal classification on each pixel in the target image feature, so that the area with pixel abnormality in the target image feature is determined as the target abnormal area corresponding to the target CT image.

[0042] In some embodiments, the use of the attention layer of the abnormality recognition model to extract pixel position-related features of the target CT image to obtain a second coding feature includes: using the image cutting network of the attention layer to cut the target CT image to obtain multiple local CT images; using the local feature extraction network of the attention layer to extract features from the multiple local CT images respectively to obtain local feature information; using the pixel association network of the attention layer to perform pixel similarity calculation on the local feature information to obtain local attention information; using the pixel association network to perform pixel similarity calculation on the target CT image to obtain global attention information; using the information splicing network of the attention layer to perform information splicing on the local attention information according to the segmentation method of the image cutting network to obtain spliced ​​attention information; using the information fusion network of the attention layer to perform weighted fusion on the global attention information and the spliced ​​attention information to obtain a fused coding feature; using the feature aggregation network of the attention layer to perform information aggregation processing on the fused coding feature to obtain the second coding feature; wherein the fused coding feature is obtained according to the following formula:

[0043]

[0044] Among them, R i represents the fused coding feature corresponding to the i-th position, G i represents the global attention information corresponding to the i-th position, L i Represents the splicing attention information corresponding to the i-th position.

[0045] Exemplarily, the attention layer includes an image segmentation network, a local feature extraction network, a pixel association network, an information splicing network, an information fusion network, and a feature aggregation network.

[0046] Exemplarily, the image cutting network obtains the number of channels, length and width corresponding to the target CT image, and then the image cutting network divides the target CT image equally according to the number of channels, length and width to obtain multiple local CT images.

[0047] Exemplarily, the local feature extraction network is a convolutional neural network, and the local feature extraction network is used to perform feature extraction on multiple local CT images respectively to obtain multiple local feature information.

[0048] Exemplarily, the pixel association network first determines a metric for similarity between pixels, such as cosine similarity, Euclidean distance, etc. This will help measure the similarity between local feature information. Thus, through the attention mechanism of the pixel association network, the extracted local feature information is used to perform similarity calculations on each pixel. This can be done by comparing the features of each pixel with the features of its surrounding pixels to obtain a similarity score. Based on the results of the similarity calculation, a weighted value is generated for each pixel. Greater weights are assigned between pixels with higher similarity, and smaller weights are assigned between pixels with lower similarity. These weights will reflect which pixels are more important in local feature extraction. The weighted values ​​are integrated into a local attention map to obtain local attention information.

[0049] Similarly, the target CT image is also input into the pixel association network to perform pixel similarity calculation to obtain the global attention information corresponding to the target CT image.

[0050] Exemplarily, the information stitching network is used to stitch the local attention information according to the segmentation method of the image cutting network to obtain the stitching attention information corresponding to the local CT image after information stitching.

[0051] Exemplarily, the information fusion network is used to weight the global attention information and the spliced ​​attention information according to the following formula to obtain the fused coding feature:

[0052]

[0053] Among them, R i represents the corresponding fusion coding feature at the i-th position, G i represents the global attention information corresponding to the i-th position, L i Represents the corresponding splicing attention information at the i-th position.

[0054] For example, the values ​​calculated by the global attention information and the spliced ​​attention information at the same position of the target CT image are not consistent. If the corresponding elements are directly added, the incorrectly calculated non-feature pixel values ​​will be more significant. However, the result calculated by fusing the global attention information and the spliced ​​attention information according to the above formula can achieve the effect of autonomously highlighting the feature area and suppressing the non-feature area.

[0055] Exemplarily, the feature aggregation network performs feature clustering on the fused coding features to obtain a clustering result, and then sums and averages the features of each cluster in the clustering result, and then corrects the feature value corresponding to each position in the cluster to the result of summing and averaging, thereby obtaining a second coding feature.

[0056] Specifically, the target CT image is cut into multiple local CT images through the image cutting network, so that each local CT image can focus on the features of a specific area, which helps the model capture subtle and important information. The pixel similarity and attention information of the local features can be calculated through the local feature extraction network and the pixel association network. This mechanism allows the model to focus on the most relevant areas in the image during the decision-making process, improving the interpretability of the model results. Then, by calculating the pixel similarity of the entire target CT image, the global attention information can be obtained. This global attention information can help improve the accuracy of local analysis, so that the position of local features in the context can be considered, and misdiagnosis caused by local features can be avoided. Through the information splicing network, the local attention information is spliced, and then the global and spliced ​​attention information are weighted and fused through the information fusion network, which can efficiently integrate local and global features and improve the expression ability of features. This fusion mechanism can promote the model to understand and utilize image information from multiple levels and perspectives. Finally, the fused coding features are aggregated through the feature aggregation network to form a more comprehensive and comprehensive feature representation, increasing the correlation between different features. Provide support for subsequent improvement of the accuracy of anomaly detection.

[0057] Step S102: determining an associated organ corresponding to the target organ, and calculating a first associated distance between the target abnormal region and the target region corresponding to the associated organ.

[0058] Exemplarily, a correlation relationship mapping table between organs is established based on expert experience or historical experience, so that the target organ is queried in the correlation relationship mapping table to obtain the associated organ corresponding to the target organ.

[0059] Exemplarily, a target image containing both the target organ and the associated organ is obtained, and then the target image is subjected to target recognition using a target recognition model to obtain a first organ region corresponding to the target organ and a target region corresponding to the associated organ. The first organ region and the second organ region corresponding to the target organ in the target CT image are then size-mapped to obtain a mapping relationship between the second organ region and the first organ region, and then the target abnormal region is mapped to the target image according to the mapping relationship to obtain a target mapping region, and then the first center position corresponding to the target mapping region and the second center position corresponding to the target region are calculated, and then the distance is calculated according to the first center position and the second center position to obtain the center distance, and then the center distance is determined as the first associated distance between the target abnormal region and the target region corresponding to the associated organ.

[0060] In some embodiments, the calculation of the first associated distance between the target abnormal area and the target area corresponding to the associated organ includes: obtaining a target image that includes both the target organ and the associated organ; converting the target abnormal area to obtain a first area in the target image, and converting the target area to obtain a second area in the target image; determining the relative position relationship between the first area and the second area; obtaining a first boundary line corresponding to the first area from the target image and a second boundary line corresponding to the second area from the target image; obtaining a first target boundary corresponding to the first area from the first boundary line based on the relative position relationship; obtaining a second target boundary corresponding to the second area from the second boundary line based on the relative position relationship; and determining the first associated distance between the target abnormal area and the target area based on the first target boundary and the second target boundary.

[0061] Exemplarily, a target image containing a target organ and its associated organs is obtained from a medical image library or real-time scanning. The target image is subjected to target recognition using a target recognition model to obtain a first organ region corresponding to the target organ, and the first organ region and the second organ region corresponding to the target organ in the target CT image are size-mapped to obtain a mapping relationship between the second organ region and the first organ region, thereby mapping the target abnormal region to the target image according to the mapping relationship to obtain the first region. Similarly, the target region is converted to the coordinate space of the target image to obtain the second region.

[0062] Exemplarily, the relative positional relationship between the target organ and the associated organ is determined based on expert experience, for example, the associated organ is above the target organ, or the associated organ is below the target organ, etc., thereby determining the relative positional relationship between the target organ and the associated organ as the relative positional relationship between the first area and the second area.

[0063] Exemplarily, the boundary points of the first area are extracted from the target image by edge detection algorithms such as Canny and Sobel. The boundary points define the outline of the target area, and the coordinates of all boundary points are obtained to form a set containing all boundary points. The minimum enclosing rectangle containing all boundary points is calculated using a geometric algorithm such as a rotating caliper algorithm. This rectangle should have the smallest area and meet the requirement of being close to the periphery of the first area. According to the vertex coordinates of the calculated minimum enclosing rectangle, a corresponding rectangular frame is drawn on the target image. This frame diagram is the minimum frame diagram. The boundary of the minimum frame diagram is thus determined as the first boundary line. Similarly, the second boundary line corresponding to the second area is obtained from the target image.

[0064] Exemplarily, the relative position relationship is the position information of the second area relative to the first area. For example, the upper left corner of the target image is determined as the coordinate origin, the upper right direction in the target image is determined as the y-axis, and the leftmost downward direction in the target image is determined as the x-axis. Then, when the relative position relationship is that the second area is above the first area, the vertical perpendicular line of the coordinate point corresponding to the minimum horizontal coordinate in the first boundary line is determined as the first target boundary, and the vertical perpendicular line of the coordinate point corresponding to the maximum horizontal coordinate in the second boundary line is determined as the second target boundary; when the relative position relationship is that the second area is below the first area, the vertical perpendicular line of the coordinate point corresponding to the maximum horizontal coordinate in the first boundary line is determined as the second target boundary. The line is determined as the first target boundary, and the vertical perpendicular line of the coordinate point corresponding to the minimum horizontal coordinate in the second boundary line is determined as the second target boundary; if the relative position relationship is that the second area is on the left side of the first area, the horizontal perpendicular line of the coordinate point corresponding to the minimum coordinate in the first boundary line is determined as the first target boundary, and the horizontal perpendicular line of the coordinate point corresponding to the maximum vertical coordinate in the second boundary line is determined as the second target boundary; if the relative position relationship is that the second area is on the right side of the first area, the horizontal perpendicular line of the coordinate point corresponding to the maximum vertical coordinate in the first boundary line is determined as the first target boundary, and the horizontal perpendicular line of the coordinate point corresponding to the minimum vertical coordinate in the second boundary line is determined as the second target boundary.

[0065] Exemplarily, a boundary distance between the first target boundary and the second target boundary is calculated, so that the boundary distance is determined as a first correlation distance between the target abnormal region and the target region.

[0066] Step S103: measuring the size of the target abnormal region to obtain a first target volume corresponding to the target abnormal region.

[0067] Exemplarily, a set of data containing the actual area and actual volume of known abnormal areas is collected and determined as training data, and then a suitable machine learning or deep learning model is selected according to the characteristics of the training data, such as linear regression, support vector machine, random forest or convolutional neural network (CNN). The machine learning or deep learning model is trained using the training data to obtain a volume prediction model. The target area is obtained by calculating the area of ​​the extracted target abnormal area according to the corresponding pixel information, and the calculated target area is input as an input variable into the trained volume prediction model. The model will output the first target volume corresponding to the target abnormal area based on the learned relationship between the area and volume.

[0068] In some embodiments, the measuring the size of the target abnormal region to obtain a first target volume corresponding to the target abnormal region includes: obtaining the CT image layer number corresponding to the target CT image and the CT image slice height corresponding to the CT image layer number; calculating the pixel area corresponding to the target abnormal region in the target CT image; determining the target slice height corresponding to the target CT image according to the CT image slice height and the pixel area; and determining the first target volume corresponding to the target CT image according to the pixel area and the target slice height.

[0069] Exemplarily, the CT image layer number corresponding to the target CT image is obtained from the CT sequence, that is, the sequence number of the target CT image in the CT sequence. The cross-sectional position of the target CT image and the adjacent CT images is obtained from the CT sequence, and the distance between the target CT image and the previous CT image is determined as the CT image slice height corresponding to the CT image layer number.

[0070] Exemplarily, the extracted target abnormal region is pixel counted, and the number of pixels is converted into pixel area according to the pixel spacing of the target abnormal region, so as to obtain the first target volume corresponding to the target CT image by multiplying the previously acquired CT image slice height and the pixel area of ​​the target abnormal region and then summing them.

[0071] In some embodiments, determining the target slice height corresponding to the target CT image based on the CT image slice height and the pixel area includes: performing data fitting according to the CT image slice number and the CT image slice height to obtain a slice height fitting function; and determining the target slice height corresponding to the target CT image based on the slice height fitting function and the CT image slice number.

[0072] Exemplarily, the CT image layer number of each target CT image is extracted from the CT sequence. The CT sequence is usually composed of a series of two-dimensional images, each image representing a slice. The number of CT image layers can be determined by the index of the image file. The target slice height refers to the physical distance between two adjacent target CT images. When the number of CT image layers is larger and the division of the CT image slice height is more detailed, the deviation between the first target volume obtained by subsequent calculation and the actual volume is smaller.

[0073] Exemplarily, the CT image layers and corresponding image slice heights of all target CT images are collected to form a data set. The collected CT image layers and corresponding image slice heights are fitted using an appropriate mathematical model (such as linear regression, polynomial regression, etc.) to obtain a slice height fitting function. The input of the function is the CT image layer number, and the output is the corresponding image slice height.

[0074] For example, in order to obtain a more accurate target slice height, it is necessary to use a slice height fitting function to predict the slice height corresponding to a non-integer number of slices. Thus, the non-integer part between the CT image slice number corresponding to the target CT image and the previous CT image slice number corresponding to the previous target CT image corresponding to the target CT image is input into the slice height fitting function to calculate the predicted slice height corresponding to the non-integer. Thus, the predicted slice height and the image slice height are summed and averaged to obtain the target slice height corresponding to the target CT image.

[0075] Step S104: determining a corresponding hardener dosage when the target patient performs hardener interventional infusion on the target abnormal area according to the first target volume and the first associated distance.

[0076] Exemplarily, previous patient treatment data are collected from clinical trials and related literature. These data should include records of different types of lesions (such as tumors, cysts, etc.), corresponding target volumes, infused sclerosant doses, and associated distances. Use statistical analysis methods (such as scatter plots, correlation analysis, etc.) to preliminarily determine the relationship between target volume, associated distance, and sclerosant dose. The analysis results can help identify the main influencing factors and provide a basis for the selection of relationship models. Confirm the specific role of target volume and associated distance in the process of sclerosant dose determination, and clarify whether their specific degree of influence on sclerosant dose is positively correlated, negatively correlated, or nonlinear. According to the characteristics of the data, linear regression, polynomial regression, machine learning algorithms, and other methods can be selected to establish a relationship model between target volume and associated distance and sclerosant dose.

[0077] Exemplarily, the established relationship model is used to perform dose prediction according to the first target volume and the first associated distance of the target patient to obtain the corresponding hardener dose of the target patient under the target abnormal area.

[0078] In some embodiments, the method of determining the corresponding sclerosant dose when the target patient performs interventional sclerosant infusion to the target abnormal area based on the first target volume and the first associated distance includes: when the first target volume is smaller than a preset volume, determining the corresponding sclerosant dose when the target patient performs interventional sclerosant infusion to the target abnormal area based on the first target volume and the first associated distance; when the first target volume is greater than or equal to the preset volume, performing preset processing on the target abnormal area, and recalculating the second target volume and the second associated distance corresponding to the target abnormal area; determining the corresponding sclerosant dose when the target patient performs interventional sclerosant infusion to the target abnormal area based on the second target volume and the second associated distance.

[0079] Exemplarily, the preset volume can be set based on expert experience or historical experience, so that when the first target volume is smaller than the preset volume, the corresponding sclerosant dosage when the target patient performs sclerosant interventional infusion on the target abnormal area can be directly determined based on the first target volume and the first associated distance.

[0080] Exemplarily, when the first target volume is greater than or equal to the preset volume, the target abnormal area needs to be treated with catheter drainage followed by hardening. After sufficient drainage (usually 24 to 48 hours of drainage until the drainage volume is significantly reduced), the hardener is injected again. That is, when the first target volume is greater than or equal to the preset volume, the target abnormal area needs to be treated with catheter drainage first, and then the target CT image of the target patient is re-obtained, and then the re-obtained target CT image is used to perform abnormality recognition using the abnormality recognition model to obtain a new target abnormal area; and the second associated distance between the new target abnormal area and the target area corresponding to the associated organ is recalculated; and the new target abnormal area is measured in size to obtain the second target volume corresponding to the new target abnormal area; and finally, the corresponding hardener dose when the target patient performs hardener interventional infusion on the target abnormal area is determined based on the second target volume and the second associated distance.

[0081] The embodiment of the present invention provides a method and system for determining the dose of sclerosant interventional infusion, the method comprising: obtaining a target CT image corresponding to a target organ of a target patient, and then using an abnormality recognition model to deeply analyze the target CT image to obtain a target abnormal area. Further, by determining the associated organ of the target organ and calculating the first associated distance between the target abnormal area and the target area corresponding to the associated organ, the spatial relationship between the target abnormal area and the target area is quantified and understood, so as to provide support for the subsequent accuracy and safety of the dose and minimize the risk to the target patient, and then the size of the target abnormal area is measured, and its first target volume can be accurately calculated. Finally, the first target volume and the first associated distance are combined for analysis to form a more comprehensive and three-dimensional treatment decision model to determine the corresponding sclerosant dose when the target patient performs sclerosant interventional infusion on the target abnormal area, so that the effective treatment of the target abnormal area can be ensured through a reasonable sclerosant dose, while avoiding potential side effects caused by excessive or insufficient doses. The problem that in the related art, although CT-guided imaging technology provides important guidance for sclerosant interventional infusion, there are still significant challenges in accurately determining the sclerosant dose is solved. The reliability and accuracy of the prediction of the corresponding sclerosant dose when the sclerosant is injected into the target abnormal area is improved, while ensuring the treatment effect and improving the safety of the target patient.

[0082] See also Figure 2 , Figure 2A system 200 for determining a dose of a sclerosant interventional infusion is provided in an embodiment of the present application. The system 200 for determining a dose of a sclerosant interventional infusion comprises an abnormality recognition module 201, a distance calculation module 202, a volume calculation module 203, and a dose determination module 204, wherein the abnormality recognition module 201 is used to obtain a target CT image corresponding to a target organ of a target patient, and perform abnormality recognition on the target CT image using an abnormality recognition model to obtain a target abnormal region corresponding to the target CT image; the distance calculation module 202 is used to determine an associated organ corresponding to the target organ, and calculate a first associated distance between the target abnormal region and the target region corresponding to the associated organ; the volume calculation module 203 is used to measure the size of the target abnormal region to obtain a first target volume corresponding to the target abnormal region; and the dose determination module 204 is used to determine a corresponding sclerosant dose when the target patient performs sclerosant interventional infusion on the target abnormal region according to the first target volume and the first associated distance.

[0083] In some implementations, the abnormality recognition module 201 performs, during the process of using the abnormality recognition model to recognize abnormalities in the target CT image and obtaining the target abnormal region corresponding to the target CT image:

[0084] Using the encoder of the abnormality recognition model to extract features of the target CT image step by step to obtain a first coding feature;

[0085] Using the attention layer of the abnormality recognition model to extract pixel position related features of the target CT image to obtain a second coding feature;

[0086] According to the decoder of the abnormality recognition model, the first coding feature and the second coding feature are used to perform feature fusion and feature restoration step by step to obtain target image features;

[0087] The target image features are pixel-classified according to the pixel classification layer of the abnormality recognition model to obtain the target abnormal area corresponding to the target CT image.

[0088] In some embodiments, the abnormality recognition module 201, in the process of extracting pixel position-related features from the target CT image using the attention layer of the abnormality recognition model to obtain the second coding feature, performs:

[0089] Using the image cutting network of the attention layer to cut the target CT image to obtain multiple local CT images;

[0090] Using the local feature extraction network of the attention layer to extract features from the plurality of local CT images respectively to obtain local feature information;

[0091] Utilizing the pixel association network of the attention layer to perform pixel similarity calculation on the local feature information to obtain local attention information;

[0092] Using the pixel association network to perform pixel similarity calculation on the target CT image to obtain global attention information;

[0093] Using the information splicing network of the attention layer to perform information splicing on the local attention information according to the segmentation method of the image segmentation network, to obtain spliced ​​attention information;

[0094] Using the information fusion network of the attention layer to weightedly fuse the global attention information and the spliced ​​attention information to obtain a fused coding feature;

[0095] Using the feature aggregation network of the attention layer to perform information aggregation processing on the fused coding feature to obtain the second coding feature;

[0096] The fusion coding feature is obtained according to the following formula:

[0097]

[0098] Among them, R i represents the fused coding feature corresponding to the i-th position, G i represents the global attention information corresponding to the i-th position, L i Represents the splicing attention information corresponding to the i-th position.

[0099] In some implementations, the distance calculation module 202, during the process of calculating the first associated distance between the target abnormal region and the target region corresponding to the associated organ, performs:

[0100] Obtaining a target image including both the target organ and the associated organ;

[0101] Converting the target abnormal region into the target image to obtain a first region, and converting the target region into the target image to obtain a second region;

[0102] determining a relative positional relationship between the first area and the second area;

[0103] Obtaining a first boundary line corresponding to the first area from the target image and obtaining a second boundary line corresponding to the second area from the target image;

[0104] Obtaining a first target boundary corresponding to the first area from the first boundary line according to the relative position relationship;

[0105] Obtaining a second target boundary corresponding to the second area from the second boundary line according to the relative position relationship;

[0106] The first associated distance between the target abnormal area and the target area is determined according to the first target boundary and the second target boundary.

[0107] In some implementations, the volume calculation module 203 performs, during the process of measuring the size of the target abnormal region to obtain the first target volume corresponding to the target abnormal region:

[0108] Obtaining the CT image layer number corresponding to the target CT image and the CT image slice height corresponding to the CT image layer number;

[0109] Calculate the pixel area corresponding to the target abnormal area in the target CT image;

[0110] Determine a target slice height corresponding to the target CT image according to the CT image slice height and the pixel area;

[0111] The first target volume corresponding to the target CT image is determined according to the pixel area and the target slice height.

[0112] In some implementations, the volume calculation module 203, in the process of determining the target slice height corresponding to the target CT image according to the slice height of the CT image and the pixel area, performs:

[0113] Performing data fitting according to the number of CT image layers and the slice height of the CT image to obtain a slice height fitting function;

[0114] The target slice height corresponding to the target CT image is determined according to the slice height fitting function and the number of CT image layers.

[0115] In some embodiments, during the process of determining the corresponding sclerosant dose when the target patient performs sclerosant interventional infusion to the target abnormal area according to the first target volume and the first associated distance, the dose determination module 204 executes:

[0116] When the first target volume is smaller than a preset volume, determining the corresponding hardener dosage when the target patient performs hardener interventional infusion on the target abnormal area according to the first target volume and the first associated distance;

[0117] When the first target volume is greater than or equal to the preset volume, the target abnormal area is subjected to preset processing, and a second target volume and a second associated distance corresponding to the target abnormal area are recalculated;

[0118] The sclerosant dosage corresponding to the target abnormal area subjected to sclerosant interventional infusion is determined according to the second target volume and the second associated distance.

[0119] In some embodiments, the system 200 for determining the dosage of a sclerosing agent for interventional infusion can be applied to a terminal device.

[0120] It should be noted that those skilled in the art can clearly understand that, for the sake of convenience and brevity of description, the specific working process of the system 200 for determining the interventional infusion dose of the sclerosant described above can refer to the corresponding process in the aforementioned embodiment of the method for determining the interventional infusion dose of the sclerosant, and will not be repeated here.

[0121] See also Figure 3 , Figure 3 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention.

[0122] like Figure 3 As shown, the terminal device 300 includes a processor 301 and a memory 302, and the processor 301 and the memory 302 are connected via a bus 303, such as an I2C (Inter-integrated Circuit) bus.

[0123] Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire terminal device. The processor 301 can be a central processing unit (CPU), and the processor 301 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0124] Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0125] Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a partial structure related to the embodiment of the present invention, and does not constitute a limitation on the terminal device to which the embodiment of the present invention is applied. The specific server may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0126] The processor is used to run a computer program stored in the memory, and implement any one of the methods for determining the interventional infusion dose of a sclerosant provided in the embodiments of the present invention when executing the computer program.

[0127] In one embodiment, the processor is used to run a computer program stored in the memory, and implements the following steps when executing the computer program:

[0128] Obtaining a target CT image corresponding to a target organ of a target patient, and performing abnormality recognition on the target CT image using an abnormality recognition model to obtain a target abnormal region corresponding to the target CT image;

[0129] Determining an associated organ corresponding to the target organ, and calculating a first associated distance between the target abnormal region and the target region corresponding to the associated organ;

[0130] Measuring the size of the target abnormal area to obtain a first target volume corresponding to the target abnormal area;

[0131] The corresponding sclerosant dosage when the target patient performs sclerosant interventional infusion on the target abnormal area is determined according to the first target volume and the first associated distance.

[0132] In some implementations, the processor 301, in the process of using the abnormality recognition model to identify abnormalities in the target CT image and obtaining a target abnormal region corresponding to the target CT image, executes:

[0133] Using the encoder of the abnormality recognition model to extract features of the target CT image step by step to obtain a first coding feature;

[0134] Using the attention layer of the abnormality recognition model to extract pixel position related features of the target CT image to obtain a second coding feature;

[0135] According to the decoder of the abnormality recognition model, the first coding feature and the second coding feature are used to perform feature fusion and feature restoration step by step to obtain target image features;

[0136] The target image features are pixel-classified according to the pixel classification layer of the abnormality recognition model to obtain the target abnormal area corresponding to the target CT image.

[0137] In some implementations, the processor 301, in the process of extracting pixel position-related features from the target CT image using the attention layer of the abnormality recognition model to obtain a second coding feature, executes:

[0138] Using the image cutting network of the attention layer to cut the target CT image to obtain multiple local CT images;

[0139] Using the local feature extraction network of the attention layer to extract features from the plurality of local CT images respectively to obtain local feature information;

[0140] Utilizing the pixel association network of the attention layer to perform pixel similarity calculation on the local feature information to obtain local attention information;

[0141] Using the pixel association network to perform pixel similarity calculation on the target CT image to obtain global attention information;

[0142] Using the information splicing network of the attention layer to perform information splicing on the local attention information according to the segmentation method of the image segmentation network, to obtain spliced ​​attention information;

[0143] Using the information fusion network of the attention layer to weightedly fuse the global attention information and the spliced ​​attention information to obtain a fused coding feature;

[0144] Using the feature aggregation network of the attention layer to perform information aggregation processing on the fused coding feature to obtain the second coding feature;

[0145] The fusion coding feature is obtained according to the following formula:

[0146]

[0147] Among them, R i represents the fused coding feature corresponding to the i-th position, G i represents the global attention information corresponding to the i-th position, L i Represents the splicing attention information corresponding to the i-th position.

[0148] In some implementations, during the process of calculating the first associated distance between the target abnormal region and the target region corresponding to the associated organ, the processor 301 executes:

[0149] Obtaining a target image including both the target organ and the associated organ;

[0150] Converting the target abnormal region into the target image to obtain a first region, and converting the target region into the target image to obtain a second region;

[0151] determining a relative positional relationship between the first area and the second area;

[0152] Obtaining a first boundary line corresponding to the first area from the target image and obtaining a second boundary line corresponding to the second area from the target image;

[0153] Obtaining a first target boundary corresponding to the first area from the first boundary line according to the relative position relationship;

[0154] Obtaining a second target boundary corresponding to the second area from the second boundary line according to the relative position relationship;

[0155] The first associated distance between the target abnormal area and the target area is determined according to the first target boundary and the second target boundary.

[0156] In some implementations, during the process of measuring the size of the target abnormal region to obtain a first target volume corresponding to the target abnormal region, the processor 301 executes:

[0157] Obtaining the CT image layer number corresponding to the target CT image and the CT image slice height corresponding to the CT image layer number;

[0158] Calculate the pixel area corresponding to the target abnormal area in the target CT image;

[0159] Determine a target slice height corresponding to the target CT image according to the CT image slice height and the pixel area;

[0160] The first target volume corresponding to the target CT image is determined according to the pixel area and the target slice height.

[0161] In some implementations, during the process of determining the target slice height corresponding to the target CT image according to the CT image slice height and the pixel area, the processor 301 executes:

[0162] Performing data fitting according to the number of CT image layers and the slice height of the CT image to obtain a slice height fitting function;

[0163] The target slice height corresponding to the target CT image is determined according to the slice height fitting function and the number of CT image layers.

[0164] In some embodiments, during the process of determining the corresponding sclerosant dosage when the target patient performs sclerosant interventional infusion to the target abnormal area according to the first target volume and the first associated distance, the processor 301 executes:

[0165] When the first target volume is smaller than a preset volume, determining the corresponding hardener dosage when the target patient performs hardener interventional infusion on the target abnormal area according to the first target volume and the first associated distance;

[0166] When the first target volume is greater than or equal to the preset volume, the target abnormal area is subjected to preset processing, and a second target volume and a second associated distance corresponding to the target abnormal area are recalculated;

[0167] The sclerosant dosage corresponding to the target abnormal area subjected to sclerosant interventional infusion is determined according to the second target volume and the second associated distance.

[0168] It should be noted that those skilled in the art can clearly understand that, for the sake of convenience and brevity of description, the specific working process of the terminal device described above can refer to the corresponding process in the aforementioned embodiment of the method for determining the interventional infusion dose of the sclerosant, and will not be repeated here.

[0169] An embodiment of the present invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any method for determining the interventional infusion dose of a sclerosant as provided in the specification of the embodiment of the present invention.

[0170] The storage medium may be an internal storage unit of the terminal device described in the foregoing embodiment, such as a hard disk or memory of the terminal device. The storage medium may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc., equipped on the terminal device.

[0171] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0172] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0173] The serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for determining the dosage of a sclerosing agent for interventional infusion, characterized in that: The method comprises: Obtaining a target CT image corresponding to a target organ of a target patient, and performing abnormality recognition on the target CT image using an abnormality recognition model to obtain a target abnormal region corresponding to the target CT image; Determining an associated organ corresponding to the target organ, and calculating a first associated distance between the target abnormal region and the target region corresponding to the associated organ; Measuring the size of the target abnormal area to obtain a first target volume corresponding to the target abnormal area; The corresponding sclerosant dosage when the target patient performs sclerosant interventional infusion on the target abnormal area is determined according to the first target volume and the first associated distance.

2. The method according to claim 1, characterized in that The step of using the abnormality recognition model to perform abnormality recognition on the target CT image to obtain a target abnormal area corresponding to the target CT image includes: Using the encoder of the abnormality recognition model to extract features of the target CT image step by step to obtain a first coding feature; Using the attention layer of the abnormality recognition model to extract pixel position related features of the target CT image to obtain a second coding feature; According to the decoder of the abnormality recognition model, the first coding feature and the second coding feature are used to perform feature fusion and feature restoration step by step to obtain target image features; The target image features are pixel-classified according to the pixel classification layer of the abnormality recognition model to obtain the target abnormal area corresponding to the target CT image.

3. The method according to claim 2, characterized in that The step of extracting pixel position-related features from the target CT image using the attention layer of the abnormality recognition model to obtain a second coding feature includes: Using the image cutting network of the attention layer to cut the target CT image to obtain multiple local CT images; Using the local feature extraction network of the attention layer to extract features from the plurality of local CT images respectively to obtain local feature information; Utilizing the pixel association network of the attention layer to perform pixel similarity calculation on the local feature information to obtain local attention information; Using the pixel association network to perform pixel similarity calculation on the target CT image to obtain global attention information; Using the information splicing network of the attention layer to perform information splicing on the local attention information according to the segmentation method of the image segmentation network, to obtain spliced ​​attention information; Using the information fusion network of the attention layer to weightedly fuse the global attention information and the spliced ​​attention information to obtain a fused coding feature; Using the feature aggregation network of the attention layer to perform information aggregation processing on the fused coding feature to obtain the second coding feature; The fusion coding feature is obtained according to the following formula: Among them, R i represents the fusion coding feature corresponding to the i-th position, G i represents the global attention information corresponding to the i-th position, L i Represents the splicing attention information corresponding to the i-th position.

4. The method according to claim 1, characterized in that: The calculating a first associated distance between the target abnormal region and the target region corresponding to the associated organ comprises: Obtaining a target image including both the target organ and the associated organ; Converting the target abnormal region into the target image to obtain a first region, and converting the target region into the target image to obtain a second region; determining a relative positional relationship between the first area and the second area; Obtaining a first boundary line corresponding to the first area from the target image and obtaining a second boundary line corresponding to the second area from the target image; Obtaining a first target boundary corresponding to the first area from the first boundary line according to the relative position relationship; Obtaining a second target boundary corresponding to the second area from the second boundary line according to the relative position relationship; The first associated distance between the target abnormal area and the target area is determined according to the first target boundary and the second target boundary.

5. The method according to claim 1, characterized in that The step of measuring the size of the target abnormal region to obtain a first target volume corresponding to the target abnormal region includes: Obtaining the CT image layer number corresponding to the target CT image and the CT image slice height corresponding to the CT image layer number; Calculate the pixel area corresponding to the target abnormal area in the target CT image; Determine a target slice height corresponding to the target CT image according to the CT image slice height and the pixel area; The first target volume corresponding to the target CT image is determined according to the pixel area and the target slice height.

6. The method according to claim 5, characterized in that The step of determining a target slice height corresponding to the target CT image according to the slice height of the CT image and the pixel area includes: Performing data fitting according to the number of CT image layers and the slice height of the CT image to obtain a slice height fitting function; The target slice height corresponding to the target CT image is determined according to the slice height fitting function and the number of CT image layers.

7. The method according to claim 1, characterized in that The step of determining, according to the first target volume and the first associated distance, a corresponding sclerosant dosage when the target patient performs sclerosant interventional infusion on the target abnormal area comprises: When the first target volume is smaller than a preset volume, determining the corresponding hardener dosage when the target patient performs hardener interventional infusion on the target abnormal area according to the first target volume and the first associated distance; When the first target volume is greater than or equal to the preset volume, the target abnormal area is subjected to preset processing, and a second target volume and a second associated distance corresponding to the target abnormal area are recalculated; The sclerosant dosage corresponding to the target abnormal area subjected to sclerosant interventional infusion is determined according to the second target volume and the second associated distance.

8. A system for determining the dosage of a sclerosant interventional infusion, characterized in that: include: An abnormality recognition module is used to obtain a target CT image corresponding to a target organ of a target patient, and use an abnormality recognition model to perform abnormality recognition on the target CT image to obtain a target abnormal area corresponding to the target CT image; a distance calculation module, used to determine an associated organ corresponding to the target organ, and calculate a first associated distance between the target abnormal region and the target region corresponding to the associated organ; A volume calculation module, used for measuring the size of the target abnormal area to obtain a first target volume corresponding to the target abnormal area; The dosage determination module is used to determine the corresponding dosage of the sclerosant when the target patient performs sclerosant interventional infusion on the target abnormal area according to the first target volume and the first associated distance.

9. A terminal device, characterized in that: The terminal device includes a processor and a memory; The memory is used to store computer programs; The processor is configured to execute the computer program and implement the method for determining the interventional infusion dose of the sclerosant according to any one of claims 1 to 7 when executing the computer program.

10. A computer storage medium for computer storage, characterized in that: The computer storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the method for determining the interventional infusion dose of a sclerosant according to any one of claims 1 to 7.

Images