An imaging system and method for cardiovascular and cerebrovascular detection

Abstract

The application discloses an imaging system and method for cardiovascular and cerebrovascular detection, and relates to the technical field of medical image processing.The system comprises the following steps: monitoring the physiological state of a patient and recording the medical history of the patient to form initial data; transmitting the initial data to a Hadoop data processing platform to perform standardization processing on the data; generating a data set; importing the data set into a random forest algorithm model to select an imaging mode; and performing pre-scanning on the data set based on the selected imaging mode.The application integrates wearable device data acquisition, Hadoop data processing, random forest imaging mode selection, deep learning automatic segmentation and feature extraction, and isolated forest anomaly detection, and solves the problem of insufficient accuracy of blood vessel segmentation and feature extraction.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to an imaging system and method for cardiovascular and cerebrovascular detection. Background Art

[0002] In recent years, diagnostic technology for cardiovascular and cerebrovascular diseases has made significant progress. Traditional imaging techniques such as CT, MRI, and ultrasound are widely used in clinical practice, providing static information about vascular structures. With the development of artificial intelligence, machine learning methods such as convolutional neural networks (CNNs) and random forest algorithms have shown great potential in vascular segmentation and lesion detection, advancing the automation level of medical image processing. However, existing technologies still have many limitations.

[0003] Traditional cardiovascular imaging methods rely on a single mode and cannot fully capture the dynamic changes and functional characteristics of blood vessels, resulting in inaccurate and incomplete diagnostic results. Data processing is concentrated on a single platform and lacks distributed processing capabilities, which affects the efficiency of large-scale real-time data analysis. Existing vascular segmentation algorithms lack accuracy when facing complex vascular structures and are prone to mis-segmentation or missed segmentation. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an imaging system and method for cardiovascular and cerebrovascular detection to solve the problem of insufficient accuracy of blood vessel segmentation and feature extraction in the prior art.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides an imaging method for cardiovascular and cerebrovascular detection, comprising:

[0008] Monitor the patient's physiological status and record the patient's medical history to form initial data;

[0009] Transfer the initial data to the Hadoop data processing platform to standardize the data and generate a data set;

[0010] Importing the dataset into the random forest algorithm model to select an imaging mode, pre-scanning the dataset based on the selected imaging mode, acquiring images of vascular structure and dynamic changes, and transmitting the images of vascular structure and dynamic changes to the distributed edge computing platform for preliminary processing;

[0011] A deep convolutional neural network model is used to automatically segment and extract features from the preliminarily processed vascular structure and dynamic change images to obtain segmented images of the blood vessels as well as their morphological, functional, and pathological characteristics.

[0012] Based on the segmented images of blood vessels and the morphological, functional and pathological characteristics of the blood vessels, the preset gradient boosting tree is used to identify abnormalities in the blood vessels, and a diagnostic report is generated based on the identification results.

[0013] As a preferred embodiment of the imaging method for cardiovascular and cerebrovascular detection of the present invention, wherein: the monitoring of the patient's physiological state and the recording of the patient's medical history to form initial data are specifically performed as follows:

[0014] Wearable devices and medical monitors are used to regularly collect patient physiological data and transmit it to a central server;

[0015] Use the BERT model to parse unstructured text data from patient physiological data in the central server, extract diagnosis information, treatment information, medical history information, examination results, and symptom description information, and convert them into a structured format;

[0016] Match real-time physiological data with medical history data by patient ID and arrange them by timestamp to form a time series dataset;

[0017] The data in the arranged time series dataset were deduplicated using Pandas, missing values ​​were filled using KNN, and outliers were removed using Z-score. The quality of the patient's physiological data was then cross-validated using K-fold to form the initial data.

[0018] As a preferred embodiment of the imaging method for cardiovascular and cerebrovascular detection of the present invention, the initial data is transmitted to the Hadoop data processing platform for data standardization to generate a data set, and the specific steps are as follows:

[0019] Export the cleaned initial data as a Parquet file;

[0020] Use StandardScaler to standardize the numeric fields in the Parquet file and convert them into distributed fields with a mean of 0 and a standard deviation of 1.

[0021] Use SparkSQL to extract time series features from the distribution field and partition the data set by patient ID.

[0022] Store the partitioned dataset in Parquet format and in an HDFS directory.

[0023] As a preferred embodiment of the imaging method for cardiovascular and cerebrovascular detection described in the present invention, the data set is imported into a random forest algorithm model to select an imaging mode, the data set is pre-scanned based on the selected imaging mode, and the vascular structure and dynamic change images are obtained and transmitted to a distributed edge computing platform for preliminary processing. The specific steps are as follows:

[0024] Load the partitioned Parquet files from the HDFS directory and use Spark to load the data, generating a SparkDataFrame of patient ID, time series features, normalized physiological data, and medical history data.

[0025] Extract feature columns from SparkDataFrame and combine them into feature vectors to generate training and test datasets;

[0026] Use the random forest classifier to build a random forest algorithm model, set the number of trees, maximum depth, and feature subset selection strategy to automatic, and substitute the training dataset into the random forest algorithm model for training to generate a trained random forest algorithm model;

[0027] Use the trained model to make predictions on the test set, generate the resulting dataset, and select the imaging mode;

[0028] Select and configure the corresponding imaging equipment according to the imaging mode predicted by the random forest model;

[0029] Pre-scanning the patient using the selected imaging device to generate a preliminary image dataset;

[0030] The preliminary image dataset is transferred to the distributed edge computing platform to perform preliminary processing such as denoising, enhancement and segmentation on the image data;

[0031] The preliminarily processed image data is stored in the edge computing platform.

[0032] As a preferred embodiment of the imaging method for cardiovascular and cerebrovascular detection described in the present invention, the method adopts a deep convolutional neural network model and automatically segments and extracts features from the preliminarily processed vascular structure and dynamic change images to obtain segmented images of the blood vessels as well as the morphological, functional, and pathological characteristics of the blood vessels. The specific steps are as follows:

[0033] Load the preliminarily processed image data from the edge computing platform and use the encoder part of the U-Net model to extract the multi-level features of the image;

[0034] The decoder part upsamples the multi-level features to generate a segmented image of the blood vessels, and then post-processes the segmented image to remove noise and artifacts;

[0035] F is used as a comprehensive score of vascular features to extract the morphological, functional, and pathological characteristics of blood vessels from the segmented image. The formula is as follows:

[0036]

[0037] Where n is the number of vascular regions, A iis the area of ​​the i-th blood vessel region, P i is the perimeter of the ith vascular region, d i is the diameter of the i-th blood vessel region, μ d is the mean diameter of the vascular area, d i is the perimeter of the i-th vascular region, σ d is the standard deviation of the diameter of the vascular area;

[0038] The calculation results are used to calculate the morphological characteristics, functional characteristics, vascular pathological characteristics and image data of abnormal vascular areas of each vascular area and store them on the edge computing platform.

[0039] As a preferred embodiment of the imaging method for cardiovascular and cerebrovascular detection described in the present invention, the method includes: identifying abnormalities of blood vessels using a preset gradient boosting tree based on the segmented image of the blood vessels and the morphological, functional and pathological characteristics of the blood vessels, and generating a diagnosis report based on the identification results, wherein the specific steps are:

[0040] The morphological, functional and lesion features of the blood vessels are fused in a weighted manner through the self-attention mechanism;

[0041] Based on the fused morphological, functional, and lesion feature vectors of the blood vessels, a graph neural network is used to detect image data of abnormal vascular areas and extract feature data of the abnormal areas.

[0042] The isolation forest model is used to calculate the anomaly score for the feature data of the abnormal area;

[0043] A diagnostic report is generated based on the abnormality score and image data of the vascular abnormality area.

[0044] As a preferred embodiment of the imaging method for cardiovascular and cerebrovascular detection of the present invention, wherein: the isolation forest model is used to calculate the abnormality score of the feature data of the abnormal area, and the specific steps are:

[0045] Use the Z-score standardization method to standardize the feature data of the abnormal area and generate a standardized feature data set;

[0046] The standardized feature dataset is trained using the isolation forest model to generate an isolation tree. The anomaly score of each data point is calculated using the path length of the isolation tree and the anomaly score is normalized.

[0047] Compare the characteristic data of the abnormal area with the lesion feature library to determine the lesion type;

[0048] A diagnostic report including the location of the abnormal area, lesion type and abnormality score is generated and output to doctors and patients.

[0049] In a second aspect, the present invention provides an imaging system for cardiovascular and cerebrovascular detection, comprising:

[0050] The monitoring module monitors the patient's physiological status and records the patient's medical history to form the initial data;

[0051] The preprocessing module transfers the initial data to the Hadoop data processing platform to standardize the data and generate a data set;

[0052] The imaging and pre-scanning module imports the dataset into the random forest algorithm model to select the imaging mode, pre-scans the dataset based on the selected imaging mode, obtains images of vascular structure and dynamic changes, and transmits the images of vascular structure and dynamic changes to the distributed edge computing platform for preliminary processing;

[0053] The image processing module uses a deep convolutional neural network model to automatically segment and extract features from the initially processed vascular structure and dynamic change images, obtaining segmented images of the blood vessels as well as their morphological, functional, and pathological characteristics.

[0054] The diagnostic analysis module identifies vascular abnormalities based on the segmented images of the blood vessels and the morphological, functional and pathological characteristics of the blood vessels through a preset gradient boosting tree, and generates a diagnostic report based on the identification results.

[0055] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the imaging method for cardiovascular and cerebrovascular detection as described in the first aspect of the present invention is implemented.

[0056] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the imaging method for cardiovascular and cerebrovascular detection as described in the first aspect of the present invention is implemented.

[0057] The beneficial effects of the present invention are as follows: by integrating wearable devices, Hadoop platform, random forest algorithm, deep convolutional neural network and isolation forest model, multimodal imaging, distributed data processing and intelligent analysis are realized. Wearable devices and medical monitors collect patient physiological data, and medical history data is analyzed in combination with BERT model to form a structured initial data set. The Hadoop platform performs data standardization processing to improve processing efficiency and scalability. The random forest algorithm intelligently selects the optimal imaging mode. The deep convolutional neural network automatically segments and extracts features, significantly improving the accuracy of blood vessel segmentation and reducing mis-segmentation and missed segmentation in complex blood vessel structures. The gradient boosting tree and isolation forest model perform anomaly detection and generate accurate diagnosis reports. The various steps work together to significantly improve the accuracy of blood vessel segmentation and feature extraction, avoiding the lack of accuracy of the blood vessel segmentation algorithm when facing complex blood vessel structures, which is prone to mis-segmentation or missed segmentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0059] Figure 1 Flowchart of the imaging method for cardiovascular and cerebrovascular detection in Example 1.

[0060] Figure 2 This is a flowchart for calculating the abnormality score in Example 1. DETAILED DESCRIPTION

[0061] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0062] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0063] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0064] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides an imaging method for cardiovascular and cerebrovascular detection, comprising the following steps:

[0065] S1. Monitor the patient's physiological status and record the patient's medical history to form the initial data.

[0066] Furthermore, wearable devices and medical monitors are used to regularly collect physiological data of patients and transmit them to a central server.

[0067] Specifically, the combination of wearable devices and medical monitors enables comprehensive and continuous collection of patients' physiological data, avoiding the discontinuous or missed data collection issues associated with traditional methods. Real-time data transmission to a central server ensures centralized management and efficient utilization of data, providing a reliable foundation for subsequent data analysis and processing.

[0068] Furthermore, the BERT model is used to parse the unstructured text data of patient physiological data in the central server, extract diagnosis information, treatment information, medical history information, examination results and symptom description information and convert them into a structured format.

[0069] Specifically, the BERT model's natural language processing capabilities enable efficient extraction of key information such as diagnosis, treatment, and medical history, and conversion of this information into a structured format. This avoids the inefficiencies and errors inherent in traditional manual processing of unstructured data. The generation of structured data provides a high-quality data source for subsequent data matching and analysis.

[0070] Furthermore, real-time physiological data are matched with medical history data by patient ID and arranged by timestamp to form a time series dataset.

[0071] Specifically, by matching patients by ID and arranging by timestamp, we can organically combine patients' real-time physiological data with historical medical data to form a complete time series dataset. This data integration method helps to fully understand the patient's health status and disease evolution, providing more comprehensive and accurate data support for subsequent data analysis and diagnosis.

[0072] Furthermore, the data in the arranged time series dataset were deduplicated using Pandas, missing values ​​were filled using KNN, and outliers were removed using Z-score. The quality of the patient physiological data was verified by K-fold cross-validation to finally form the initial data.

[0073] Specifically, Pandas deduplication, KNN imputation of missing values, and Z-score removal of outliers effectively addressed noise and anomalies in the data, ensuring data accuracy and consistency. K-fold cross-validation further verified data quality, avoiding the issues of incomplete data cleaning or insufficient validation found in traditional methods. The resulting initial dataset was high-quality and reliable, providing a solid foundation for subsequent machine learning and diagnostic analysis.

[0074] It should be noted that through real-time data acquisition, intelligent analysis, data integration and time series processing, data cleaning, and quality verification, comprehensive and efficient processing of patient physiological and medical history data is achieved. Each step optimizes the shortcomings of traditional methods to ensure data integrity, accuracy, and usability. The resulting initial dataset provides high-quality data support for subsequent imaging mode selection, image processing, and diagnostic analysis, significantly improving the accuracy and efficiency of cardiovascular and cerebrovascular testing.

[0075] S2. The initial data is transferred to the Hadoop data processing platform to standardize the data and generate a data set.

[0076] Furthermore, the cleaned initial data is exported as a Parquet file.

[0077] Specifically, using the Parquet file format significantly reduces data storage space and transmission time while maintaining the data's columnar storage structure, facilitating efficient subsequent querying and analysis. Compared to traditional CSV or JSON formats, Parquet files offer higher performance and lower storage costs when processing large amounts of data, providing a sound data foundation for subsequent distributed processing.

[0078] Furthermore, StandardScaler is used to standardize the numeric fields in the Parquet file and convert them into distributed fields with a mean of 0 and a standard deviation of 1.

[0079] Specifically, normalization can eliminate dimensional differences between features, preventing excessively large values ​​in some features from unnecessarily impacting model training. Normalized data better meets the input requirements of machine learning algorithms, improving model training efficiency and prediction accuracy. This can significantly enhance model stability and generalization capabilities, especially when analyzing multiple features.

[0080] Furthermore, SparkSQL is used to extract time series features from the distribution field and the dataset is partitioned and stored by patient ID.

[0081] Specifically, by extracting time series features, we can capture trends and patterns in patient physiological data over time, providing important feature input for subsequent time series analysis and prediction. Partitioning datasets by patient ID improves data query efficiency and management convenience. Partitioned storage can significantly reduce data retrieval time and improve overall system performance, especially when processing large-scale data.

[0082] Furthermore, the partitioned dataset is stored in Parquet format in an HDFS directory.

[0083] Specifically, storing data in HDFS enables distributed data management and high availability, ensuring system stability and scalability in large-scale data processing scenarios. HDFS's distributed storage mechanism effectively copes with growing data volumes, while the Parquet columnar storage structure further improves data query efficiency and processing performance. This storage approach provides strong support for subsequent distributed computing and real-time analysis.

[0084] It should be noted that comprehensive optimization and management of the initial data was achieved through efficient data storage, standardized processing, time series feature extraction, and distributed storage. Each step addressed the shortcomings of traditional methods, ensuring data efficiency, comparability, and manageability. The resulting dataset provided high-quality data input for subsequent random forest algorithm model training and imaging mode selection, significantly improving the overall performance and accuracy of the cardiovascular and cerebrovascular monitoring system.

[0085] S3. Import the dataset into the random forest algorithm model and select the imaging mode. Pre-scan the dataset based on the selected imaging mode to obtain the vascular structure and dynamic change images and transmit the vascular structure and dynamic change images to the distributed edge computing platform for preliminary processing.

[0086] Furthermore, the partitioned Parquet files are loaded from the HDFS directory and Spark is used to load the data to generate SparkDataFrames of patient ID, time series features, standardized physiological data, and medical history data.

[0087] Specifically, using Spark to load Parquet files leverages the advantages of distributed computing, significantly improving data loading and processing speed. The structured data format of Spark DataFrame facilitates subsequent feature extraction and model training, avoiding the inefficient data loading associated with traditional methods and providing a foundation for efficient data processing for subsequent machine learning models.

[0088] Furthermore, feature columns are extracted from SparkDataFrame and combined into feature vectors to generate training and test datasets.

[0089] Specifically, by extracting feature vectors and partitioning the dataset, we ensure data independence and representativeness for model training and testing. Generating feature vectors enables machine learning models to fully utilize multidimensional features for training, avoiding the problems of insufficient feature selection or irrational dataset partitioning in traditional methods, and improving model training effectiveness and generalization capabilities.

[0090] Furthermore, a random forest algorithm model is constructed using a random forest classifier, the number of trees, maximum depth, and feature subset selection strategy are set to automatic, and the training dataset is substituted into the random forest algorithm model for training to generate a trained random forest algorithm model.

[0091] Specifically, the random forest algorithm's multi-tree ensemble and automatic feature subset selection effectively improves the model's classification accuracy and robustness. The random forest algorithm excels at handling high-dimensional data and complex feature relationships, avoiding the overfitting problem often associated with traditional single-decision tree models. The trained model accurately captures complex patterns in the data, providing a reliable predictive foundation for subsequent imaging modality selection.

[0092] Furthermore, the trained model is used to make predictions on the test set, generate the result dataset and select the imaging mode.

[0093] Specifically, through model prediction and imaging mode selection, the most appropriate imaging mode can be dynamically selected based on the patient's physiological data and historical records, avoiding the problems of single or inaccurate imaging mode selection in traditional methods. Intelligent imaging mode selection significantly improves imaging targeting and diagnostic accuracy, providing an optimal imaging solution for subsequent image acquisition and processing.

[0094] It should be noted that through efficient data loading, feature vector generation, random forest model training, and intelligent imaging mode selection, a comprehensive intelligent processing process, from data loading to imaging mode selection, has been achieved. Each step addresses the shortcomings of traditional methods, ensuring efficient data processing, accurate model training, and intelligent imaging mode selection. The resulting imaging mode provides an optimal solution for subsequent image acquisition and diagnostic analysis, significantly improving the overall performance and diagnostic accuracy of the cardiovascular and cerebrovascular detection system.

[0095] Furthermore, the corresponding imaging device is selected and configured according to the imaging mode predicted by the random forest model.

[0096] Specifically, the random forest model's predictions dynamically select the most appropriate imaging mode based on the patient's physiological data and historical records, avoiding the limitations of traditional methods, which often rely on a single or inaccurate imaging mode selection. Intelligent imaging mode selection significantly improves imaging targeting and diagnostic accuracy, providing an optimal imaging solution for subsequent image acquisition and processing.

[0097] Furthermore, a pre-scan of the patient is performed using the selected imaging device to generate a preliminary image dataset.

[0098] Specifically, generating a preliminary image dataset through pre-scanning allows for rapid acquisition of image information about vascular structure and dynamic changes, avoiding the incomplete or low-quality image acquisition issues inherent in traditional methods. This preliminary image dataset provides a high-quality data foundation for subsequent image processing and analysis, significantly improving both efficiency and accuracy.

[0099] Furthermore, the preliminary image data set is transmitted to the distributed edge computing platform for preliminary processing of image data such as denoising, enhancement and segmentation.

[0100] Specifically, the distributed edge computing platform enables real-time processing close to the data source, reducing data transmission latency and improving the efficiency and real-time nature of image processing. Preliminary processing operations such as denoising, enhancement, and segmentation significantly improve the quality of image data, providing high-quality image input for subsequent deep learning models and avoiding the inefficient and low-quality image processing often associated with traditional methods.

[0101] Furthermore, the preliminarily processed image data is stored in the edge computing platform.

[0102] Specifically, storing pre-processed image data on the edge computing platform enables efficient data management and rapid access, ensuring system stability and scalability in large-scale data processing scenarios. The distributed storage mechanism of the edge computing platform effectively copes with data growth while improving data query efficiency and processing performance, providing strong support for subsequent deep learning model training and diagnostic analysis.

[0103] It should be noted that through intelligent imaging mode selection, pre-scan image acquisition, distributed edge computing platform preliminary processing, and image data storage, a full intelligent processing process from imaging mode selection to preliminary image data processing is achieved. Each step is optimized to address the shortcomings of traditional methods, ensuring the accuracy of imaging modes, high-quality image acquisition, real-time image processing, and efficient data storage. The resulting preliminary processed image data provides a high-quality data foundation for subsequent deep learning model training and diagnostic analysis, significantly improving the overall performance and diagnostic accuracy of the cardiovascular and cerebrovascular detection system.

[0104] S4. A deep convolutional neural network model is used to automatically segment and extract features from the preliminarily processed vascular structure and dynamic change images to obtain segmented images of the blood vessels as well as the morphological, functional, and pathological characteristics of the blood vessels.

[0105] Furthermore, the preliminarily processed image data is loaded from the edge computing platform and the encoder part of the U-Net model is used to extract the multi-level features of the image.

[0106] Specifically, the encoder portion of the U-Net model extracts multi-layered features from the image, capturing detailed information about vascular structures and avoiding the problem of insufficient feature extraction in traditional methods. This multi-layered feature extraction provides a rich information foundation for subsequent image segmentation and feature analysis, significantly improving segmentation and feature extraction accuracy.

[0107] Furthermore, the multi-level features are upsampled through the decoder part to generate a segmented image of the blood vessels, and the segmented image is post-processed to remove noise and artifacts.

[0108] Specifically, upsampling and post-processing in the decoder generate high-quality segmented images, removing noise and artifacts, thus avoiding the low-quality segmented images often encountered in traditional methods. These high-quality segmented images provide a reliable data foundation for subsequent feature extraction and lesion detection, significantly improving the accuracy and reliability of lesion detection.

[0109] Furthermore, F is used as a comprehensive score of vascular features to extract the morphological, functional, and pathological characteristics of blood vessels from the segmented image. The formula is as follows:

[0110]

[0111] Where n is the number of vascular regions, A i is the area of ​​the i-th blood vessel region, P i is the perimeter of the ith vascular region, d i is the diameter of the i-th blood vessel region, μ d is the mean diameter of the vascular area, d i is the perimeter of the i-th vascular region, σ d is the standard deviation of the vessel area diameter.

[0112] Specifically, a comprehensive scoring formula enables quantitative assessment of vascular morphology, functionality, and lesion characteristics, avoiding the subjectivity inherent in traditional feature assessment methods. This quantitative assessment provides objective data support for subsequent lesion detection and diagnosis, significantly improving the accuracy and reliability of lesion detection.

[0113] Furthermore, the morphological characteristics, functional characteristics, vascular pathological characteristics and image data of abnormal vascular areas of each vascular area are calculated through the calculation results and stored in the edge computing platform.

[0114] Specifically, by storing vascular characteristics on the edge computing platform, efficient data management and rapid access are achieved, ensuring system stability and scalability in large-scale data processing scenarios. The distributed storage mechanism of the edge computing platform can effectively cope with the growth of data volume, while improving data query efficiency and processing performance, providing strong support for subsequent lesion detection and diagnostic analysis.

[0115] It should be noted that accurate segmentation and feature extraction of vascular images were achieved through a deep convolutional neural network model, multi-level feature extraction, image segmentation and post-processing, quantitative evaluation of vascular features, and data storage. Each step was optimized to address the shortcomings of traditional methods, ensuring accurate image segmentation, comprehensive feature extraction, and efficient data storage. The resulting vascular features provide high-quality data support for subsequent lesion detection and diagnostic analysis, significantly improving the overall performance and diagnostic accuracy of the cardiovascular and cerebrovascular detection system.

[0116] S5. Based on the segmented image of the blood vessels and the morphological, functional and pathological characteristics of the blood vessels, the abnormal conditions of the blood vessels are identified through a preset gradient boosting tree, and a diagnosis report is generated based on the identification results.

[0117] Furthermore, the morphological, functional and pathological features of the blood vessels are fused in a weighted manner through the self-attention mechanism.

[0118] Specifically, the self-attention mechanism enables weighted fusion of different features based on their importance, avoiding the issues of insufficient feature fusion or irrational weight distribution found in traditional methods. The weighted fusion of feature vectors can more comprehensively reflect vascular abnormalities, significantly improving the accuracy and reliability of subsequent anomaly detection.

[0119] Furthermore, based on the fused morphological, functional and lesion feature vectors of the blood vessels, a graph neural network is used to detect image data of abnormal vascular areas and extract feature data of the abnormal areas.

[0120] Specifically, the powerful graph structure analysis capabilities of graph neural networks can effectively capture the complex relationships between abnormal vascular regions, avoiding the inaccurate abnormal region detection problems encountered by traditional methods. Graph neural networks can process the topological information of vascular structures, significantly improving the accuracy of abnormal region detection and the comprehensiveness of feature extraction, providing high-quality data support for subsequent anomaly score calculations.

[0121] Furthermore, the isolation forest model is used to calculate the anomaly score for the feature data of the abnormal area.

[0122] Specifically, the Isolation Forest model can effectively identify outliers within the characteristic data of abnormal regions, avoiding the subjective nature of anomaly scoring in traditional methods. The Isolation Forest model excels at processing high-dimensional data, accurately calculating anomaly scores and significantly improving the objectivity and accuracy of abnormal region assessments, providing a reliable quantitative basis for subsequent diagnostic report generation.

[0123] Furthermore, a diagnosis report is generated based on the abnormality score and the image data of the abnormal vascular area.

[0124] Specifically, through comprehensive analysis of abnormality scores and test results, a detailed diagnostic report can be generated, avoiding the inefficient or incomplete reporting issues of traditional methods. The diagnostic report provides doctors with clear lesion information and quantitative assessment results, significantly improving diagnostic efficiency and accuracy and providing a reliable basis for patient treatment.

[0125] It should be noted that intelligent identification and diagnosis of vascular anomalies is achieved through gradient boosting tree anomaly recognition, self-attention mechanism feature fusion, graph neural network abnormal region detection, isolation forest anomaly score calculation, and diagnostic report generation. Each step optimizes the shortcomings of traditional methods to ensure the accuracy of anomaly recognition, the comprehensiveness of feature fusion, the precision of abnormal region detection, the objectivity of anomaly scoring, and the efficiency of diagnostic reports. The resulting diagnostic report provides doctors with a reliable basis for diagnosis, significantly improving the overall performance and diagnostic accuracy of the cardiovascular and cerebrovascular detection system.

[0126] S6. Use the isolation forest model to calculate the anomaly score for the feature data of the abnormal area.

[0127] Furthermore, the Z-score standardization method is used to standardize the feature data of the abnormal area to generate a standardized feature data set.

[0128] Specifically, Z-score normalization eliminates dimensional differences between features, preventing excessively large values ​​of certain features from unnecessarily impacting model training. Normalized data better meets the input requirements of the Isolation Forest model, improving model training efficiency and prediction accuracy. This process can significantly enhance model stability and generalization, especially when analyzing multiple features.

[0129] Furthermore, the isolation forest model is used to train the standardized feature dataset to generate an isolation tree. The anomaly score of each data point is calculated by the path length of the isolation tree and the anomaly score is normalized.

[0130] Specifically, through the training of the isolation forest model and the calculation of path lengths, we can accurately assess the degree of abnormality of each data point, avoiding the subjective nature of anomaly scoring in traditional methods. The normalized anomaly scores make the results more intuitive and easier to understand, providing a reliable basis for subsequent lesion type determination and diagnostic report generation.

[0131] Furthermore, the characteristic data of the abnormal area is compared with the lesion characteristic library to determine the lesion type.

[0132] Specifically, by comparing it with a lesion signature database, the lesion type in the abnormal area can be accurately identified, avoiding the inaccurate lesion type determination problem encountered in traditional methods. Intelligent lesion type determination provides accurate data support for subsequent diagnostic report generation, significantly improving the accuracy and reliability of lesion detection.

[0133] Furthermore, a diagnostic report including the location of the abnormal area, lesion type and abnormality score is generated and output to doctors and patients.

[0134] Specifically, the automated generation of diagnostic reports quickly provides detailed diagnostic information, including the location of abnormal areas, lesion type, and abnormality score, eliminating the inefficiency of traditional diagnostic report generation methods. The automated output of diagnostic reports provides doctors and patients with timely and accurate diagnostic evidence, significantly improving diagnostic efficiency and accuracy.

[0135] It should be noted that intelligent detection and diagnosis of abnormal areas is achieved through the standardization of feature data, calculation and normalization of anomaly scores, lesion type determination, and diagnostic report generation. Each step addresses the shortcomings of traditional methods, ensuring the accuracy of anomaly detection, precise lesion type determination, and timely diagnostic report generation. The resulting diagnostic report provides doctors and patients with a reliable diagnostic basis, significantly improving the overall performance and diagnostic accuracy of the cardiovascular and cerebrovascular detection system.

[0136] This embodiment also provides a spectral axis control adjustment system for assisting in myopia prevention and control, including:

[0137] The monitoring module monitors the patient's physiological status and records the patient's medical history to form the initial data;

[0138] The preprocessing module transfers the initial data to the Hadoop data processing platform to standardize the data and generate a data set;

[0139] The imaging and pre-scanning module imports the dataset into the random forest algorithm model to select the imaging mode, pre-scans the dataset based on the selected imaging mode, obtains images of vascular structure and dynamic changes, and transmits the images of vascular structure and dynamic changes to the distributed edge computing platform for preliminary processing;

[0140] The image processing module uses a deep convolutional neural network model to automatically segment and extract features from the initially processed vascular structure and dynamic change images, obtaining segmented images of the blood vessels as well as their morphological, functional, and pathological characteristics.

[0141] The diagnostic analysis module identifies vascular abnormalities based on the segmented images of the blood vessels and the morphological, functional and pathological characteristics of the blood vessels through a preset gradient boosting tree, and generates a diagnostic report based on the identification results.

[0142] This embodiment also provides a computer device suitable for the imaging method for cardiovascular and cerebrovascular detection, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the imaging method for cardiovascular and cerebrovascular detection proposed in the above embodiment.

[0143] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0144] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the imaging method for cardiovascular and cerebrovascular detection proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0145] In summary, the present invention realizes multimodal imaging, distributed data processing and intelligent analysis by integrating wearable devices, Hadoop platform, random forest algorithm, deep convolutional neural network and isolation forest model. Wearable devices and medical monitors collect patient physiological data, and medical history data is analyzed in combination with BERT model to form a structured initial data set. The Hadoop platform performs data standardization processing to improve processing efficiency and scalability. The random forest algorithm intelligently selects the optimal imaging mode. The deep convolutional neural network automatically segments and extracts features, significantly improving the accuracy of blood vessel segmentation and reducing mis-segmentation and missed segmentation in complex blood vessel structures. The gradient boosting tree and isolation forest model perform anomaly detection and generate accurate diagnosis reports. Each step works together to significantly improve the accuracy of blood vessel segmentation and feature extraction, avoiding the insufficient accuracy of the blood vessel segmentation algorithm when facing complex blood vessel structures, which is prone to mis-segmentation or missed segmentation.

[0146] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An imaging method for cardiovascular and cerebrovascular detection, characterized by: include, Monitor the patient's physiological status and record the patient's medical history to form the initial data. Match the real-time physiological data with the medical history data by patient ID and arrange them by timestamp to form a time series data set. Transfer the initial data to the Hadoop data processing platform to standardize the data and generate a data set; Importing the dataset into the random forest algorithm model to select an imaging mode, pre-scanning the dataset based on the selected imaging mode, acquiring images of vascular structure and dynamic changes, and transmitting the images of vascular structure and dynamic changes to the distributed edge computing platform for preliminary processing; A deep convolutional neural network model is used to automatically segment and extract features from the preliminarily processed vascular structure and dynamic change images to obtain segmented images of the blood vessels as well as their morphological, functional, and pathological characteristics. Based on the segmented images of blood vessels and the morphological, functional and pathological characteristics of the blood vessels, the system uses a preset gradient boosting tree to identify abnormalities in the blood vessels and generates a diagnostic report based on the identification results. The data set is imported into the random forest algorithm model to select an imaging mode, the data set is pre-scanned based on the selected imaging mode, the vascular structure and dynamic change images are obtained, and the vascular structure and dynamic change images are transmitted to the distributed edge computing platform for preliminary processing. The specific steps are: Load the partitioned Parquet files from the HDFS directory and use Spark to load the data, generating a SparkDataFrame of patient ID, time series features, normalized physiological data, and medical history data. Extract feature columns from SparkDataFrame and combine them into feature vectors to generate training and test datasets; Use the random forest classifier to build a random forest algorithm model, set the number of trees, maximum depth, and feature subset selection strategy to automatic, and substitute the training dataset into the random forest algorithm model for training to generate a trained random forest algorithm model; Use the trained model to make predictions on the test set, generate the resulting dataset, and select the imaging mode; Select and configure the corresponding imaging equipment according to the imaging mode predicted by the random forest model; Pre-scanning the patient using the selected imaging device to generate a preliminary image dataset; The preliminary image dataset is transferred to the distributed edge computing platform to perform preliminary processing such as denoising, enhancement and segmentation on the image data; The preliminarily processed image data is stored in the edge computing platform.

2. The imaging method for cardiovascular and cerebrovascular detection according to claim 1, wherein: The monitoring of the patient's physiological state and recording of the patient's medical history to form the initial data are specifically carried out as follows: Wearable devices and medical monitors are used to regularly collect patient physiological data and transmit it to a central server; Use the BERT model to parse unstructured text data from patient physiological data in the central server, extract diagnosis information, treatment information, medical history information, examination results, and symptom description information, and convert them into a structured format; The data in the arranged time series dataset were deduplicated using Pandas, missing values ​​were filled using KNN, and outliers were removed using Z-score. The quality of the patient's physiological data was then cross-validated using K-fold to form the initial data.

3. The imaging method for cardiovascular and cerebrovascular detection according to claim 2, wherein: The initial data is transferred to the Hadoop data processing platform to perform data standardization and generate a data set. The specific steps are: Export the cleaned initial data as a Parquet file; Use StandardScaler to standardize the numeric fields in the Parquet file and convert them into distributed fields with a mean of 0 and a standard deviation of 1. Use SparkSQL to extract time series features from the distribution field and partition the data set by patient ID. Store the partitioned dataset in Parquet format and in an HDFS directory.

4. The imaging method for cardiovascular and cerebrovascular detection according to claim 1, wherein: The method uses a deep convolutional neural network model to automatically segment and extract features from the preliminarily processed vascular structure and dynamic change images to obtain vascular segmentation images as well as vascular morphology, functionality, and pathological characteristics. The specific steps are: Load the preliminarily processed image data from the edge computing platform and use the encoder part of the U-Net model to extract the multi-level features of the image; The decoder part upsamples the multi-level features to generate a segmented image of the blood vessels, and then post-processes the segmented image to remove noise and artifacts; References As a comprehensive score of vascular features, the morphological, functional and pathological characteristics of the blood vessels are extracted from the segmented image. The formula is as follows: ; in, is the number of vascular regions, For the The area of ​​the vascular region, For the The perimeter of the vascular area, For the The diameter of the vascular area, is the mean diameter of the vascular area, is the standard deviation of the diameter of the vascular area; By calculating the results The morphological characteristics, functional characteristics, vascular pathological characteristics and image data of abnormal vascular areas of each vascular region are calculated and stored on the edge computing platform.

5. The imaging method for cardiovascular and cerebrovascular detection according to claim 1, wherein: The method is to identify abnormalities of blood vessels using a preset gradient boosting tree based on the segmented image of the blood vessels and the morphological, functional and pathological characteristics of the blood vessels, and generate a diagnosis report based on the identification results. The specific steps are: The morphological, functional and lesion features of the blood vessels are fused in a weighted manner through the self-attention mechanism; Based on the fused morphological, functional, and lesion feature vectors of the blood vessels, a graph neural network is used to detect image data of abnormal vascular areas and extract feature data of the abnormal areas. The isolation forest model is used to calculate the anomaly score for the feature data of the abnormal area; A diagnostic report is generated based on the abnormality score and image data of the vascular abnormality area.

6. The imaging method for cardiovascular and cerebrovascular detection according to claim 5, wherein: The isolation forest model is used to calculate the abnormality score of the feature data of the abnormal area, and the specific steps are: Use the Z-score standardization method to standardize the feature data of the abnormal area and generate a standardized feature data set; The standardized feature dataset is trained using the isolation forest model to generate an isolation tree. The anomaly score of each data point is calculated using the path length of the isolation tree and the anomaly score is normalized. Compare the characteristic data of the abnormal area with the lesion feature library to determine the lesion type; A diagnostic report including the location of the abnormal area, lesion type and abnormality score is generated and output to doctors and patients.

7. An imaging system for cardiovascular and cerebrovascular disease detection, based on the imaging method for cardiovascular and cerebrovascular disease detection according to any one of claims 1 to 6, characterized in that: include, The monitoring module monitors the patient's physiological status and records the patient's medical history to form the initial data; The preprocessing module transfers the initial data to the Hadoop data processing platform to standardize the data and generate a data set; The imaging and pre-scanning module imports the dataset into the random forest algorithm model to select the imaging mode, pre-scans the dataset based on the selected imaging mode, obtains images of vascular structure and dynamic changes, and transmits the images of vascular structure and dynamic changes to the distributed edge computing platform for preliminary processing; The image processing module uses a deep convolutional neural network model to automatically segment and extract features from the initially processed vascular structure and dynamic change images, obtaining segmented images of the blood vessels as well as their morphological, functional, and pathological characteristics. The diagnostic analysis module identifies vascular abnormalities based on the segmented images of the blood vessels and the morphological, functional and pathological characteristics of the blood vessels through a preset gradient boosting tree, and generates a diagnostic report based on the identification results.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the imaging method for cardiovascular and cerebrovascular detection according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the imaging method for cardiovascular and cerebrovascular detection according to any one of claims 1 to 6 are implemented.

Images