Free-breathing coronary scan image lesion analysis system for the elderly

Abstract

The application discloses an old person free breathing coronary artery scanning image lesion analysis system and relates to the technical field of scanning image lesion analysis. In order to solve the problem that the accurate condition of a patient cannot be obtained according to a scanning image. The application adopts a diameter method, an area method and a contrast agent filling condition to judge the stenosis degree and the hemodynamic change, analyzes the lesion from the morphological and functional double angles, provides comprehensive information for clinical decision-making, identifies the R wave peak value and divides the cardiac phase, matches the respiratory signal and the projection data in time, can accurately capture the characteristics of the heart in different motion states and the respiratory stage, effectively avoids the interference of the artifacts caused by the heart beat and the respiratory motion, makes the reconstructed image clearer and more accurate, sets the CT and the injector parameters in sequence from the scanning type confirmation, each link is closely related and the target is clear, improves the work efficiency, and is convenient for quality control and process management.

Description

Technical Field

[0001] This invention relates to the field of lesion analysis technology in scanned images, specifically a lesion analysis system for coronary artery scanned images of elderly individuals during free breathing. Background Technology

[0002] Scan image lesion analysis is a method that uses medical imaging technology to identify, assess, and diagnose lesion areas.

[0003] Chinese patent CN118648025A discloses an apparatus and method for reconstructing three-dimensional oral cavity scan data using computed tomography (CT) images. It primarily generates three-dimensional coordinate information and three-dimensional feature points from geometrically distortion-free CT images, and locally matches scan keyframes acquired by the scanner with the CT images to reconstruct the three-dimensional model. This reduces errors caused by the cumulative matching of scan keyframes, thereby lowering the geometric distortion of the oral cavity scan model. While this patent solves the image reconstruction problem, the following issues remain in practical operation:

[0004] 1. The lack of accurate provision and adjustment of patient information and equipment parameters resulted in poor overall scanning results.

[0005] 2. The patient's scan images were not further reconstructed and optimized, resulting in unclear scan images.

[0006] 3. The lack of targeted image evaluation of the patient's scan results in the inability to obtain timely information about the patient's lesions. Summary of the Invention

[0007] The purpose of this invention is to provide a coronary artery lesion analysis system for elderly patients with free breathing. It employs diameter and area methods combined with contrast agent filling to determine the degree of stenosis and hemodynamic changes, analyzing lesions from both morphological and functional perspectives to provide comprehensive information for clinical decision-making. By identifying R-wave peak values ​​and dividing cardiac phases, and combining the time matching of respiratory signals and projection data, it can accurately capture the characteristics of the heart in different motion states and respiratory stages, effectively avoiding artifact interference caused by cardiac pulsation and respiratory motion, resulting in clearer and more accurate reconstructed images. From scan type confirmation to sequential setting of CT and syringe parameters, each step is closely linked and has a clear objective, improving work efficiency and facilitating quality control and process management, thus solving problems in existing technologies.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A system for analyzing coronary artery lesions in elderly individuals during free breathing, including:

[0010] First, patient information is collected from the database. Then, the parameters of the third-generation dual-source CT and high-pressure dual-barrel injector are set according to the scanning requirements. During the patient scan, scanning data is collected in real time. The real-time collected scanning data is reconstructed and scanned to generate scanned images. The generated scanned images are then optimized. The image quality of the optimized scanned images is assessed, and lesion analysis is performed on the scanned images based on the quality assessment results.

[0011] Preferably, patient information is collected from a database, including:

[0012] Before collecting patient information, the staff member's identity must be verified. After the database verification is successful, the staff member can log in to the database to collect patient information.

[0013] Patient information includes basic information and clinical information;

[0014] The basic information includes the patient's name, the department visited, the doctor who visited, and the patient's identification; the clinical information includes the examination request form, medical history, and laboratory test results.

[0015] Based on the collected patient information, staff members conduct on-site verification with the patients.

[0016] If the verification is correct, the patient will proceed with the scan; if there is an error, staff will adjust the missing or abnormal information.

[0017] Preferably, the parameters of the third-generation dual-source CT and the high-pressure dual-barrel injector are set according to the scanning requirements, including:

[0018] The scan requirements are retrieved from the rule base. The scan type is confirmed based on the scan requirements and patient information. The scan type includes coronary artery enhancement scan or calcium scoring scan.

[0019] After confirming the scan type, set the parameters for the third-generation dual-source CT and the high-pressure dual-barrel injector in sequence;

[0020] The parameter setting steps for the third-generation dual-source CT are as follows:

[0021] Use retrospective ECG-gated spiral scanning or prospective ECG-triggered axial scanning, and enable respiratory exercise compensation technology;

[0022] The scanning range extends from the level of the tracheal carina to 1 cm below the diaphragm, covering the origin of the coronary arteries to their distal branches;

[0023] Next, the X-ray parameters are set, including tube voltage and tube current. The tube voltage is 100kV or 120kV. 100kV is selected for patients weighing <70kg or elderly patients, and 120kV is selected for patients weighing ≥70kg or with severe calcification. The tube current adopts automatic tube current adjustment technology, which is automatically adjusted according to the patient's body size and scanning area. Elderly patients with low BMI should manually reduce the base tube current.

[0024] The pitch and layer thickness are set as follows: the pitch is calculated based on the heart rate and is set to 0.2-0.3 mm; the layer thickness is 0.5-0.75 mm; and the interlayer spacing is 0.5 mm.

[0025] The parameter setting steps for a high-pressure dual-barrel injector are as follows:

[0026] Choose isotonic or hypotonic nonionic contrast agents with a concentration of 350-400 mgI / mL. The contrast agent dosage is calculated based on the patient's weight.

[0027] The injection rate is 5.0-6.0 mL / s, reduced to 4.5-5.0 mL / s for the elderly; the saline flushing volume is 30-50 mL; and the injection pressure is ≤300 PSI.

[0028] Finally, the parameter settings for the third-generation dual-source CT and high-pressure dual-barrel injector were completed.

[0029] Preferably, during the patient scan, scan data is acquired in real time, including:

[0030] First, guide the patient into the scanning room and connect them to vital signs monitoring equipment;

[0031] Start the third-generation dual-source CT to perform a localization image scan, acquire the patient's chest anteroposterior and lateral projection images, adjust the scanning range according to the localization image, and check again whether the CT parameter settings meet the scanning requirements.

[0032] Once the criteria are met, a formal scan and contrast agent injection will be performed. The injection program will be started according to preset parameters. First, the contrast agent will be injected, and then the system will automatically switch to saline flushing. After the contrast agent injection begins, the staff will trigger the scan command on the CT console and simultaneously start ECG gating synchronization and respiratory exercise compensation.

[0033] ECG-gated synchronization involves continuously recording ECG signals during retrospective ECG-gated spiral scanning and matching the scan data with cardiac motion; respiratory motion compensation involves enabling respiratory gating or motion tracking technology on the CT device to record respiratory amplitude in real time.

[0034] The CT tube and detector rotate around the patient's chest, acquiring projection data layer by layer, while simultaneously recording synchronous data, including electrocardiogram signals, respiratory signals, gantry selection angle, radiation dose, and scan time.

[0035] Finally, projection data, synchronization data, and device operation data are obtained, while real-time acquisition of scanning data is completed.

[0036] Preferably, the real-time acquired scan data is reconstructed to generate a scanned image, including:

[0037] After the scan is completed, the projection data, synchronization data and equipment operation data collected in real time by the CT equipment are automatically transmitted to the image reconstruction workstation.

[0038] The image reconstruction workstation processes the received data, including identifying the peak value of each R wave in the electrocardiogram signal, dividing the heart into 10 equal phases, matching the respiratory signal with the time error of the projected data, and identifying the respiratory phase.

[0039] After the image reconstruction workstation completes the data processing, it selects a reconstruction algorithm, which includes conventional reconstruction and ECG-gated reconstruction.

[0040] After selecting the reconstruction algorithm, the reconstruction parameters are set, including layer thickness, interlayer spacing, field of view, and reconstruction matrix;

[0041] The image reconstruction workstation performs three-dimensional reconstruction of the projection data based on the reconstruction algorithm and reconstruction parameters, generating several original cross-sectional images. Staff can then use the dynamic browsing function to select an image of the end-diastolic phase of the heart from these original cross-sectional images as the primary diagnostic scan image.

[0042] Preferably, the generated scanned image undergoes image optimization processing, including:

[0043] The scanned images are manually or automatically cropped according to the scanning range. At the same time, the window width and window level are adjusted according to the characteristics of coronary artery imaging to highlight the contrast between the vascular structure and the surrounding soft tissue, fat, and calcifications.

[0044] The scanned images are then corrected and denoised. Correction involves using the respiratory signals recorded during scanning to perform phase alignment on image misalignment caused by respiratory movements, and using ECG gating data to perform multiphase image fusion or interpolation on blurred vascular areas caused by heartbeats. Denoising involves using unsharpened filtering or anisotropic diffusion algorithms to reduce granular artifacts caused by radiation dose and equipment noise during scanning, while preserving vascular edge details.

[0045] After the scanned images are corrected and denoised, image enhancement is performed, including using gradient-based edge enhancement algorithms or vascular tensor filtering to highlight the boundaries of the coronary artery lumen. At the same time, histogram adjustment is used to amplify the gray-level difference between the blood vessels and plaques. In addition, partial volume effect caused by layer thickness is compensated through interpolation of adjacent layer images or three-dimensional convolution processing.

[0046] Finally, the enhanced scan images are reconstructed into coronal, sagittal, and curved images at arbitrary angles, unfolded into planar images along the course of the coronary arteries, and a two-dimensional MIP image of the vascular tree is generated to highlight the contrast-filled luminal structure.

[0047] Finally, the optimized processing of the scanned images is completed.

[0048] Preferably, adjusting the window width and window level according to the characteristics of coronary artery imaging includes:

[0049] The contrast between the vascular structures in the current scan image and the surrounding soft tissue, fat, and calcifications is compared and evaluated, and a pre-selected adjustment rule is determined based on the evaluation results.

[0050] Based on whether the patient has a history of coronary atherosclerosis, select the appropriate screening criteria, and obtain the estimated CT range of the coronary arteries, as well as the first window width and first window level, from the preset parameter mapping table;

[0051] Based on the patient's actual disease and disease development stage, a correlation level analysis was performed with the coronary artery imaging quality. The actual disease with a correlation level of not low correlation was considered as the disease in the correlation analysis.

[0052] If no related diseases are found in the current patient's medical history, the first window width or first window level is fine-tuned based on the pre-selected adjustment rules, and then output as the adjusted window width and adjusted window level.

[0053] If the current patient's medical history contains associated diseases, the predicted CT range for the corresponding diseases is determined based on the disease development stage of the associated diseases.

[0054] Obtain the predicted CT range of the associated disease and the predicted CT range of the coronary artery;

[0055] If there is only a single estimated CT midpoint, then the second window width and second window level are determined based on the current estimated CT midpoint.

[0056] If there are multiple estimated CT intermediate ranges, the upper and lower limits of each estimated CT intermediate range are extracted, sorted, and the median value is taken as the comprehensive CT intermediate range. Then, the second window width and the second window level are determined based on the comprehensive CT intermediate range.

[0057] By incorporating disease correlation, a comprehensive analysis of the first window width and first window level, as well as the second window width and second window level, is conducted to obtain the new window width and new window level.

[0058] The new window width or position is fine-tuned using pre-selected adjustment rules and then output as the adjusted window width and position.

[0059] Preferably, the image quality of the optimized scanned image is evaluated, including:

[0060] The optimized scanned images undergo preliminary quality screening, including the detection of motion artifacts, ray noise, and equipment artifacts. After the detection is completed, a contrast evaluation is performed.

[0061] After the initial quality screening is completed, the vascular structure display is evaluated, including the branching of blood vessels and the assessment of edge clarity.

[0062] After the vascular structure assessment is completed, the effects on respiratory movement and cardiac pulsation are assessed.

[0063] The preliminary quality screening, vascular structure visualization assessment, and assessment of the effects of respiratory movement and cardiac pulsation were obtained and then quantitatively evaluated.

[0064] Based on the quantitative evaluation results, the optimized scanned images are classified into quality levels, including acceptable quality, minor abnormalities, and severe abnormalities.

[0065] Preferably, lesion analysis is performed on the scanned images based on the quality assessment results, including:

[0066] Based on the image quality assessment results, scanned images that fail the assessment will be optimized or rescanned.

[0067] Analyze lesions using scanned images that have passed the image quality assessment;

[0068] The lesion analysis is as follows:

[0069] The system automatically identifies the main trunk and major branches of the coronary arteries in scanned images using deep learning models or traditional image processing algorithms, and marks the identified main trunk and major branches of the coronary arteries as suspicious lesion areas.

[0070] Plaques in areas of suspicion are classified into plaque types, including calcified plaques, non-calcified plaques, and mixed plaques.

[0071] Next, the degree of stenosis in the suspected lesion area is assessed, including measuring the ratio of the inner diameter or cross-sectional area of ​​the stenotic vessel to the proximal normal vessel using the diameter method or area method, and calculating the stenosis impact assessment score in conjunction with the contrast agent filling status shown by angiography.

[0072] Based on the stenosis impact assessment score, it is determined whether the stenosis leads to hemodynamic changes;

[0073] High-risk features are identified based on plaque classification and stenosis, and a visual report is generated from the identified high-risk features.

[0074] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0075] 1. The coronary artery scan image lesion analysis system for elderly patients with free breathing provided by this invention uses identity authentication, allowing only authorized personnel to log in to the database. This reduces the risk of patient information leakage at the source and protects patient privacy. From confirming the scan type to setting the CT and syringe parameters sequentially, each step is closely linked and has a clear objective, which improves work efficiency and facilitates quality control and process management.

[0076] 2. The coronary artery scan image lesion analysis system for elderly patients with free breathing provided by this invention highlights the boundaries of the coronary artery lumen through gradient-based edge enhancement algorithms or vascular tensor filtering. It also combines histogram adjustment to amplify the grayscale difference between blood vessels and plaques, while compensating for the partial volume effect caused by slice thickness. This makes details such as vascular structure and plaque features clearer and more obvious. By identifying the R-wave peak value and dividing the cardiac phase, and combining the time matching of respiratory signals and projection data, it can accurately capture the characteristics of the heart in different motion states and respiratory stages, effectively avoiding artifact interference caused by cardiac pulsation and respiratory motion, and making the reconstructed image clearer and more accurate.

[0077] 3. The coronary artery scan image lesion analysis system for elderly patients with free breathing provided by this invention classifies plaques and determines the degree of stenosis in suspicious lesion areas, achieving multi-dimensional quantitative assessment. This helps doctors understand the nature of the lesion. The system uses diameter and area methods combined with contrast agent filling to determine the degree of stenosis and hemodynamic changes, analyzing the lesion from both morphological and functional perspectives. This provides comprehensive information for clinical decision-making. By quantifying and comparing various assessment results and classifying them into acceptable quality, minor abnormalities, and severe abnormalities, the system transforms image quality assessment from subjective judgment to objective data support. Attached Figure Description

[0078] Figure 1 This is a schematic diagram of the steps for analyzing lesions in a free-breathing coronary artery scan image according to the present invention. Detailed Implementation

[0079] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0080] To address the issue in existing technologies where patient information and equipment parameters are not accurately provided and adjusted before scanning, resulting in poor overall scan quality, please refer to [link to relevant documentation]. Figure 1This embodiment provides the following technical solution:

[0081] A system for analyzing coronary artery lesions in elderly individuals during free breathing, including:

[0082] First, patient information is collected from the database. Then, the parameters of the third-generation dual-source CT and high-pressure dual-barrel injector are set according to the scanning requirements. During the patient scan, scanning data is collected in real time. The real-time collected scanning data is reconstructed and scanned to generate scanned images. The generated scanned images are then optimized. The image quality of the optimized scanned images is assessed, and lesion analysis is performed on the scanned images based on the quality assessment results.

[0083] Specifically, patient information is automatically collected from the database, avoiding errors and time-consuming issues associated with manual entry, ensuring the accuracy of basic patient data, and providing a reliable basis for subsequent scanning parameter adjustments. A free-breathing scanning mode is adopted, utilizing the high temporal resolution of dual-source CT to complete data acquisition while the patient is breathing naturally, avoiding heart rate abnormalities or respiratory artifacts caused by breath-holding. Through iterative reconstruction algorithms and noise suppression technology, the original scan data is finely reconstructed, improving the clarity of coronary artery vessel edges by 40% and increasing the display rate of small branches to 92%. A deep learning-based computer-aided diagnostic module can automatically identify lesion characteristics such as coronary artery stenosis, plaque nature, and vascular remodeling, generating a quantitative report containing parameters such as lesion location, degree of stenosis, and plaque burden within 10 seconds.

[0084] Patient information is collected from the database, including:

[0085] Before collecting patient information, the staff member's identity must be verified. After successful verification, the staff member can log into the database to collect patient information.

[0086] Patient information includes basic information and clinical information;

[0087] The basic information includes the patient's name, the department visited, the doctor who visited, and the patient's identification; the clinical information includes the examination request form, medical history, and laboratory test results.

[0088] Based on the collected patient information, staff members conduct on-site verification with the patients.

[0089] If the verification is correct, the patient will proceed with the scan; if there is an error, staff will adjust the missing or abnormal information.

[0090] Specifically, staff identity verification is used as a prerequisite for information collection, effectively preventing unauthorized access to the database. Through identity verification, only authorized staff can log in to the database, reducing the risk of patient information leakage at the source, protecting patient privacy, and maintaining the standardization and seriousness of medical institution data management. Patient information is collected through detailed classification of basic and clinical information, covering key data in the patient's diagnosis and treatment process. After collection, staff are required to verify the information with the patient on-site. This dual mechanism of "collection + verification" can promptly identify and correct errors and missing information. Adjustments can be made if errors are found during verification, ensuring the authenticity and reliability of the patient information entered into the database, providing accurate data support for subsequent diagnosis, treatment, and research.

[0091] The parameters of the third-generation dual-source CT and the high-pressure dual-barrel injector are set according to the scanning requirements, including:

[0092] The scan requirements are retrieved from the rule base. The scan type is confirmed based on the scan requirements and patient information. The scan type includes coronary artery enhancement scan or calcium scoring scan.

[0093] After confirming the scan type, set the parameters for the third-generation dual-source CT and the high-pressure dual-barrel injector in sequence;

[0094] The parameter setting steps for the third-generation dual-source CT are as follows:

[0095] Use retrospective ECG-gated spiral scanning or prospective ECG-triggered axial scanning, and enable respiratory exercise compensation technology;

[0096] The scanning range extends from the level of the tracheal carina to 1 cm below the diaphragm, covering the origin of the coronary arteries to their distal branches;

[0097] Next, the X-ray parameters are set, including tube voltage and tube current. The tube voltage is 100kV or 120kV. 100kV is selected for patients weighing <70kg or elderly patients, and 120kV is selected for patients weighing ≥70kg or with severe calcification. The tube current adopts automatic tube current adjustment technology, which is automatically adjusted according to the patient's body size and scanning area. Elderly patients with low BMI should manually reduce the base tube current.

[0098] The pitch and layer thickness are set as follows: the pitch is calculated based on the heart rate and is set to 0.2-0.3 mm; the layer thickness is 0.5-0.75 mm; and the interlayer spacing is 0.5 mm.

[0099] The parameter setting steps for a high-pressure dual-barrel injector are as follows:

[0100] Choose isotonic or hypotonic nonionic contrast agents with a concentration of 350-400 mgI / mL. The contrast agent dosage is calculated based on the patient's weight.

[0101] The injection rate is 5.0-6.0 mL / s, reduced to 4.5-5.0 mL / s for the elderly; the saline flushing volume is 30-50 mL; and the injection pressure is ≤300 PSI.

[0102] Finally, the parameter settings for the third-generation dual-source CT and high-pressure dual-barrel injector were completed.

[0103] Specifically, the scanning requirements are retrieved from the rule base and the scan type is confirmed in conjunction with patient information. This ensures that the scanning operation follows industry standards and clinical norms while allowing for flexible adjustments based on individual patient conditions, avoiding a "one-size-fits-all" approach. The parameter settings of the third-generation dual-source CT fully consider the patient's physical characteristics. Tube voltage is selected based on weight, age, and degree of calcification. Automatic tube current adjustment technology combined with manual fine-tuning reduces patient radiation dose while ensuring image quality, providing special protection for vulnerable groups such as the elderly. Pitch and slice thickness are personalized based on factors such as heart rate, optimizing scanning imaging effects and improving the detection rate of subtle lesions. The parameter settings of the high-pressure dual-barrel injector focus on the safe and effective use of contrast agents. The contrast agent dosage is calculated based on the patient's weight, and the injection rate is adjusted according to age to balance the contrast agent enhancement effect with patient tolerance. Clear limits on saline flushing dosage and injection pressure reduce the risk of adverse reactions to contrast agents, ensuring patient safety while ensuring uniform distribution of contrast agents within the blood vessels, improving scanning image quality. Clear parameter setting steps enable operators to follow standard procedures, reducing operational errors. From confirming the scan type to setting the CT and syringe parameters sequentially, each step is closely linked and has a clear objective. This not only improves work efficiency but also facilitates quality control and process management, providing clear guidance for medical team collaboration and ensuring the smooth conduct of CT scans.

[0104] To address the issue of unclear scan images resulting from the lack of further image reconstruction and optimization in existing technologies, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution:

[0105] During the patient scan, scan data is collected in real time, including:

[0106] First, guide the patient into the scanning room and connect them to vital signs monitoring equipment;

[0107] Start the third-generation dual-source CT to perform a localization image scan, acquire the patient's chest anteroposterior and lateral projection images, adjust the scanning range according to the localization image, and check again whether the CT parameter settings meet the scanning requirements.

[0108] Once the criteria are met, a formal scan and contrast agent injection will be performed. The injection program will be started according to preset parameters. First, the contrast agent will be injected, and then the system will automatically switch to saline flushing. After the contrast agent injection begins, the staff will trigger the scan command on the CT console and simultaneously start ECG gating synchronization and respiratory exercise compensation.

[0109] ECG-gated synchronization involves continuously recording ECG signals during retrospective ECG-gated spiral scanning and matching the scan data with cardiac motion; respiratory motion compensation involves enabling respiratory gating or motion tracking technology on the CT device to record respiratory amplitude in real time.

[0110] The CT tube and detector rotate around the patient's chest, acquiring projection data layer by layer, while simultaneously recording synchronous data, including electrocardiogram signals, respiratory signals, gantry selection angle, radiation dose, and scan time.

[0111] Finally, projection data, synchronization data, and device operation data are obtained, while real-time acquisition of scanning data is completed.

[0112] Specifically, positioning scans acquire anteroposterior and lateral chest projection images, allowing for precise coverage of lesion areas by adjusting the scan range accordingly. Retrospective ECG-gated spiral scanning matches scan data with cardiac motion, while respiratory gating or motion tracking technology records respiratory amplitude in real time, effectively eliminating cardiac pulsation and respiratory motion artifacts, resulting in clearer images. This provides rich and accurate information for disease diagnosis, reducing missed and misdiagnosed rates. Connecting life monitoring equipment before scanning allows for real-time monitoring of patient vital signs, timely detection and intervention of abnormalities, ensuring a safe scanning process. Contrast agent injection employs an automated program with preset parameters, injecting the contrast agent first and then automatically switching to saline flushing to ensure rational contrast agent use and reduce the risk of adverse reactions. Simultaneous activation of ECG gating and respiratory motion compensation reduces repeated scans, lowers the radiation dose received by the patient, and protects their health. The CT tube and detector automatically rotate around the patient's chest to acquire projection data, simultaneously recording multiple synchronous data such as ECG and respiration, as well as equipment operation data, achieving efficient and real-time acquisition of multi-source data. This rich and comprehensive data not only meets clinical diagnostic needs but also provides strong support for medical image analysis, equipment performance evaluation, and other scientific research, promoting the development of the medical field.

[0113] The real-time acquired scan data is reconstructed to generate scanned images, including:

[0114] After the scan is completed, the projection data, synchronization data and equipment operation data collected in real time by the CT equipment are automatically transmitted to the image reconstruction workstation.

[0115] The image reconstruction workstation processes the received data, including identifying the peak value of each R wave in the electrocardiogram signal, dividing the heart into 10 equal phases, matching the respiratory signal with the time error of the projected data, and identifying the respiratory phase.

[0116] After the image reconstruction workstation completes the data processing, it selects a reconstruction algorithm, which includes conventional reconstruction and ECG-gated reconstruction.

[0117] After selecting the reconstruction algorithm, the reconstruction parameters are set, including layer thickness, interlayer spacing, field of view, and reconstruction matrix;

[0118] The image reconstruction workstation performs three-dimensional reconstruction of the projection data based on the reconstruction algorithm and reconstruction parameters, generating several original cross-sectional images. Staff can then use the dynamic browsing function to select an image of the end-diastolic phase of the heart from these original cross-sectional images as the primary diagnostic scan image.

[0119] Specifically, after the scan, the CT equipment automatically transmits the projection data, synchronization data, and equipment operation data to the image reconstruction workstation, avoiding data omissions or transmission delays caused by manual intervention. This process achieves seamless integration of data acquisition and transmission, allowing raw data to quickly enter the processing stage. This lays the foundation for efficient image reconstruction and shortens the time patients wait for diagnostic results. The image reconstruction workstation's precise processing of ECG and respiratory signals is a highlight of the solution. By identifying R-wave peaks and dividing cardiac phases, combined with the time matching of respiratory signals and projection data, the characteristics of the heart in different motion states and respiratory stages can be accurately captured. This effectively avoids artifact interference caused by cardiac pulsation and respiratory motion, making the reconstructed images clearer and more accurate, providing reliable evidence for doctors' diagnoses. Combined with flexible settings of multiple parameters such as slice thickness and slice spacing, personalized image reconstruction solutions can be customized according to individual patient differences and clinical diagnostic needs. For example, for patients with arrhythmia, the ECG-gated reconstruction algorithm can better freeze cardiac motion and obtain clear cardiac images; while conventional reconstruction is suitable for general cases, improving diagnostic efficiency. After generating multiple cross-sectional raw images, staff can intuitively compare images of different phases through dynamic browsing. Among numerous images, images from the end-diastolic phase of the heart are precisely selected as the primary diagnostic images. During this period, the heart is in a relatively static state, and the vascular morphology is most clearly displayed, which helps doctors to more accurately determine the condition of coronary artery lesions and improve the accuracy and reliability of diagnosis.

[0120] The generated scanned images undergo image optimization processing, including:

[0121] The scanned images are manually or automatically cropped according to the scanning range. At the same time, the window width and window level are adjusted according to the characteristics of coronary artery imaging to highlight the contrast between the vascular structure and the surrounding soft tissue, fat, and calcifications.

[0122] The scanned images are then corrected and denoised. Correction involves using the respiratory signals recorded during scanning to perform phase alignment on image misalignment caused by respiratory movements, and using ECG gating data to perform multiphase image fusion or interpolation on blurred vascular areas caused by heartbeats. Denoising involves using unsharpened filtering or anisotropic diffusion algorithms to reduce granular artifacts caused by radiation dose and equipment noise during scanning, while preserving vascular edge details.

[0123] After the scanned images are corrected and denoised, image enhancement is performed, including using gradient-based edge enhancement algorithms or vascular tensor filtering to highlight the boundaries of the coronary artery lumen. At the same time, histogram adjustment is used to amplify the gray-level difference between the blood vessels and plaques. In addition, partial volume effect caused by layer thickness is compensated through interpolation of adjacent layer images or three-dimensional convolution processing.

[0124] Finally, the enhanced scan images are reconstructed into coronal, sagittal, and curved images at arbitrary angles, unfolded into planar images along the course of the coronary arteries, and a two-dimensional MIP image of the vascular tree is generated to highlight the contrast-filled luminal structure.

[0125] Finally, the optimized processing of the scanned images is completed.

[0126] Specifically, manual or automatic cropping can be flexibly selected based on the scanning range, and the window width and level can be adjusted according to the characteristics of coronary artery imaging. This allows for the rapid removal of redundant information and precise highlighting of the contrast between vascular structures and surrounding tissues. This not only reduces interference from invalid image data but also allows doctors to observe the condition of the coronary arteries more intuitively, significantly improving diagnostic efficiency. Image correction using respiratory signals and ECG gating data effectively solves the problems of image misalignment and blurring caused by respiratory movements and heartbeats, ensuring image integrity and accuracy. At the same time, denoising using unsharpened filtering or anisotropic diffusion algorithms reduces artifacts caused by equipment noise and radiation dose while preserving vascular edge details to the greatest extent, avoiding the loss of key information due to excessive denoising. This provides a clear and reliable image foundation for diagnosis. Gradient-based edge enhancement algorithms or vascular tensor filtering highlight the boundaries of the coronary artery lumen, and histogram adjustment amplifies the grayscale difference between vessels and plaques, while compensating for some volumetric effects caused by slice thickness, making details such as vascular structures and plaque features clearer and more obvious. This comprehensive image enhancement process helps doctors more accurately assess vascular lesions and improve diagnostic precision. The optimized images are reconstructed into coronal, sagittal, and curved surface images at arbitrary angles, and a two-dimensional MIP image of the vascular tree is generated. This unfolds along the coronary arteries from multiple perspectives into a planar image, comprehensively and intuitively displaying the morphology, course, and lesion locations of the coronary arteries. This rich image presentation allows doctors to observe and analyze from different angles, providing a more comprehensive and accurate basis for developing treatment plans.

[0127] Adjusting window width and window level according to the characteristics of coronary artery imaging includes:

[0128] The contrast between the vascular structures in the current scan image and the surrounding soft tissue, fat, and calcifications is compared and evaluated, and a pre-selected adjustment rule is determined based on the evaluation results.

[0129] Based on whether the patient has a history of coronary atherosclerosis, select the appropriate screening criteria, and obtain the estimated CT range of the coronary arteries, as well as the first window width and first window level, from the preset parameter mapping table;

[0130] Based on the patient's actual disease and disease development stage, a correlation level analysis was performed with the coronary artery imaging quality. The actual disease with a correlation level of not low correlation was considered as the disease in the correlation analysis.

[0131] If no related diseases are found in the current patient's medical history, the first window width or first window level is fine-tuned based on the pre-selected adjustment rules, and then output as the adjusted window width and adjusted window level.

[0132] If the current patient's medical history contains associated diseases, the predicted CT range for the corresponding diseases is determined based on the disease development stage of the associated diseases.

[0133] Obtain the predicted CT range of the associated disease and the predicted CT range of the coronary artery;

[0134] If there is only a single estimated CT midpoint, then the second window width and second window level are determined based on the current estimated CT midpoint.

[0135] If there are multiple estimated CT intermediate ranges, the upper and lower limits of each estimated CT intermediate range are extracted, sorted, and the median value is taken as the comprehensive CT intermediate range. Then, the second window width and the second window level are determined based on the comprehensive CT intermediate range.

[0136] By incorporating disease correlation, a comprehensive analysis of the first window width and first window level, as well as the second window width and second window level, is conducted to obtain the new window width and new window level.

[0137] The new window width or position is fine-tuned using pre-selected adjustment rules and then output as the adjusted window width and position.

[0138] In this embodiment, the evaluation result refers to the conclusion drawn after comparing and evaluating the contrast between the vascular structure and the surrounding soft tissue, fat, and calcifications in the current scanned image, such as blood vessels being too dark, insufficient contrast, or excessive contrast. Contrast refers to the degree of grayscale difference between the vascular structure and the surrounding soft tissue, fat, and calcifications. The pre-selected adjustment rule is an adjustment strategy for the window width or window level (consisting of specific adjustment magnitude and direction) obtained by matching the evaluation result from a preset rule library. For example, the window level is lowered by 5 units when the blood vessels are too dark; when the contrast is insufficient, the window width is increased. Different rules target different contrast problems (such as blood vessels being too dark, insufficient contrast, etc.). The preset rule library consists of a large number of window width or window level adjustment strategies based on different contrast evaluation results. These strategies are derived from clinical practice and experimental research, and are summarized from experience in adjusting images under a large number of different contrast conditions.

[0139] In this embodiment, coronary atherosclerosis refers to a disease in which lipid deposition, fibrosis, and calcium deposition occur in the intima of the coronary arteries, forming atherosclerotic plaques that lead to stenosis or blockage of the blood vessels; the preset parameter mapping table is a database that stores the estimated CT range, first window width, and first window level of the coronary arteries corresponding to different screening conditions, derived from a large amount of clinical data and experimental summaries.

[0140] In this embodiment, if the patient has a history of coronary atherosclerosis, the estimated CT range of the coronary arteries, as well as the first window width and the first window level, are obtained from the preset parameter mapping table using the first screening criteria (specifically including the patient's age, gender, and key reference indicators (specifically referring to physiological indicators related to coronary atherosclerosis, such as blood lipid levels, blood pressure levels, blood sugar levels, etc.)).

[0141] If the patient has no history of coronary atherosclerosis, the estimated CT range of the coronary arteries, as well as the first window width and first window level, are obtained from the preset parameter mapping table using the second screening criteria (specifically including the patient's age, gender, and routine scanning requirements).

[0142] In this embodiment, window width is an important parameter in CT image display technology, which defines the grayscale range of CT value display; window level refers to the center CT value of the window width, which determines the grayscale center position of the image display; the estimated CT range of coronary arteries refers to the expected CT value range of coronary arteries on the CT image obtained by matching from a preset parameter mapping table according to the patient's screening conditions; the first window width and the first window level are the initial window width and window level values ​​obtained by matching from a preset parameter mapping table according to the patient's screening conditions.

[0143] In this embodiment, the actual disease refers to the disease that the patient currently suffers from, excluding coronary atherosclerosis; the disease development stage refers to the stage of the patient's actual disease development process, generally referring to the early, middle, and late stages; the correlation level analysis refers to the level analysis and evaluation of the correlation between the patient's actual disease and the quality of coronary artery imaging. Specifically, it refers to selecting the corresponding correlation level from the set correlation level list using the name of the actual disease and the current disease development stage as matching conditions. The set correlation level list is established based on professional knowledge and experience and consists of a series of actual disease names, disease development stages, and corresponding correlation levels. For example, for lung diseases (such as emphysema), the correlation level between the disease and the quality of coronary artery imaging is set at different development stages (early, middle, and late), which may be low correlation in the early stage, medium correlation in the middle stage, and high correlation in the late stage.

[0144] In this embodiment, the disease in the association analysis refers to the actual disease that, after association level analysis, has a non-low association level with the coronary artery imaging quality. These diseases may affect the distribution of CT values ​​of the coronary arteries, such as lung diseases (e.g., emphysema, pulmonary fibrosis) and valvular heart disease. The association level is divided into three levels: low association, medium association, and high association. The disease prediction CT range refers to the predicted CT value range of the current actual disease determined based on the disease development stage and clinical experience.

[0145] In this embodiment, the estimated CT intermediate range refers to the average of the estimated CT range of the associated disease and the estimated CT range of the coronary artery. The second window width is determined based on the estimated CT intermediate range (when there is only a single estimated CT intermediate range) or the comprehensive CT intermediate range (when there are multiple estimated CT intermediate ranges), and is usually the upper limit minus the lower limit of the estimated CT intermediate range or the comprehensive CT intermediate range. The second window level is determined based on the estimated CT intermediate range (when there is only a single estimated CT intermediate range) or the comprehensive CT intermediate range (when there are multiple estimated CT intermediate ranges), and is usually the median value of the estimated CT intermediate range or the comprehensive CT intermediate range as the second window level. For example, if the comprehensive CT intermediate range is 100-200 HU, then the second window level can be set to 150 HU.

[0146] In this embodiment, the comprehensive CT intermediate range refers to the comprehensive CT intermediate range when there are multiple estimated CT intermediate ranges. The upper and lower limits of each estimated CT intermediate range are extracted, sorted, and the median value is taken as the comprehensive CT intermediate range. For example, if there are two estimated CT intermediate ranges of 80-120HU and 100-140HU, after sorting all the upper limits (120, 140) and lower limits (80, 100), the median value of the upper limit (130HU) and the median value of the lower limit (90HU) are taken, and the comprehensive CT intermediate range is 90-130HU. The new window level is obtained by introducing disease correlation and comprehensively analyzing the first window level and the second window level.

[0147] In this embodiment, for example, if there is only a single estimated CT midpoint range, and the association level of the disease corresponding to the current estimated CT midpoint range in the association analysis is medium, then a first window width exists. and the first window position Second window width and the first window position ;

[0148] At this moment, the new window is wide New window position In the formula, This represents the weight of the influence of the patient's physical condition on the analysis window width or window level. ; This represents the association weight corresponding to the association level. Different association levels correspond to different association weight values, and the value range is [value range missing]. For example, the weight for medium correlation is set to 0.4, and the weight for high correlation is set to 0.7.

[0149] In this embodiment, for example, if there are estimated CT intermediate ranges 1, 2, and 3, the upper and lower limits of the estimated CT intermediate ranges 1, 2, and 3 are extracted, sorted, and the median value is taken to obtain the comprehensive CT intermediate range 4. The estimated CT intermediate range 2 is closest to the upper and lower limits of the comprehensive CT intermediate range 4. The association level of the disease corresponding to the estimated CT intermediate range 2 in the association analysis is high, and a first window width exists. and the first window position Second window width and the first window position ;

[0150] At this moment, the new window is wide New window position In the formula, This represents the weight of the influence of the patient's physical condition on the analysis window width or window level. ; This represents the association weight corresponding to a high association level.

[0151] In this embodiment, adjusting the window width or adjusting the window level refers to the final window width or window level value used for image display obtained after fine-tuning the new window width or window level using a pre-selected adjustment rule. For example, if the window level is 100HU and the window width is 300HU, after adjusting the window level by 5 units (the unit refers to the adjustment range, i.e., 5HU) using pre-selected adjustment rule 1, the adjusted window level is 75HU, and the window width of 300HU is output as the adjusted window width.

[0152] The beneficial effects of the above technical solution are as follows: by determining the pre-selected adjustment rules based on contrast assessment, it is possible to make targeted adjustments to the contrast between vascular structures and surrounding tissues in the image; by setting and adjusting the window width and window level according to the patient's medical history and physical condition, it fully considers the individual differences of patients and the impact of other diseases on coronary artery imaging, making the window width and window level adjustment more in line with the actual condition, further optimizing image quality, and helping to provide higher quality and more accurate coronary artery imaging, thereby helping doctors to more accurately judge the condition and formulate more reasonable treatment plans.

[0153] The working principle of the above technical solution is as follows: First, the contrast between the vascular structure and surrounding tissue in the scanned image is evaluated, and pre-selected adjustment rules are matched from a preset rule library. Next, based on whether the patient has a history of coronary atherosclerosis, the estimated CT range of the coronary arteries, the first window width, and the first window level are obtained from a preset parameter mapping table using different screening conditions. Then, a correlation level analysis is performed on the patient's actual disease and its development stage to determine the associated diseases. If no associated diseases are found, the first window width or the first window level is fine-tuned according to the pre-selected adjustment rules. If associated diseases are found, the estimated CT range of the corresponding disease is determined according to the disease development stage, and the estimated CT intermediate range is obtained by comparing it with the estimated CT range of the coronary arteries. Based on this, the second window width and the second window level are determined. Afterward, disease correlation is introduced, and the first window width, second window width, first window level, and second window level are comprehensively analyzed to obtain a new window width and a new window level. Finally, the adjusted window width and the adjusted window level are output after fine-tuning using the pre-selected adjustment rules, thereby optimizing the coronary artery imaging quality and providing clearer and more accurate image evidence for clinical diagnosis.

[0154] To address the issue in existing technologies where there is no targeted image assessment of the patient's scan, resulting in the inability to obtain timely information about the patient's lesions, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution:

[0155] Image quality assessment is performed on the optimized scanned images, including:

[0156] The optimized scanned images undergo preliminary quality screening, including the detection of motion artifacts, ray noise, and equipment artifacts. After the detection is completed, a contrast evaluation is performed.

[0157] After the initial quality screening is completed, the vascular structure display is evaluated, including the branching of blood vessels and the assessment of edge clarity.

[0158] After the vascular structure assessment is completed, the effects on respiratory movement and cardiac pulsation are assessed.

[0159] The preliminary quality screening, vascular structure visualization assessment, and assessment of the effects of respiratory movement and cardiac pulsation were obtained and then quantitatively evaluated.

[0160] Based on the quantitative evaluation results, the optimized scanned images are classified into quality levels, including acceptable quality, minor abnormalities, and severe abnormalities.

[0161] Specifically, a layered and progressive evaluation process is adopted, starting with initial quality screening to detect equipment artifacts such as motion artifacts and X-ray noise, followed by contrast evaluation, and then in-depth evaluation of vascular structure display and the impact of respiration and cardiac pulsation, forming a complete detection chain. This systematic screening can comprehensively identify potential image problems, avoid missing single indicators, ensure the reliability of images used for diagnosis, and provide clinicians with solid imaging evidence. By quantifying and comparing the results of various evaluations and classifying them into acceptable quality, minor abnormalities, and severe abnormalities, image quality assessment is transformed from subjective judgment to objective data support. On the one hand, quantitative standards unify the evaluation scale, reduce evaluation differences caused by human factors, and improve the consistency of evaluation results among different doctors and in different scenarios; on the other hand, clear classification facilitates rapid classification and management of images, allowing doctors to prioritize the processing of images with severe abnormalities, improving diagnostic efficiency, and focusing on evaluating key indicators such as vascular structure display and the impact of respiration and cardiac pulsation, accurately meeting clinical diagnostic needs. Clear visualization of vascular branches and assessment of edge clarity help identify vascular lesions; assessment of the effects of respiration and cardiac pulsation can avoid the risk of misdiagnosis caused by motion interference, allowing doctors to focus more on lesion feature analysis, thereby improving the accuracy of disease diagnosis.

[0162] Based on the quality assessment results, lesion analysis was performed on the scanned images, including:

[0163] Based on the image quality assessment results, scanned images that fail the assessment will be optimized or rescanned.

[0164] Analyze lesions using scanned images that have passed the image quality assessment;

[0165] The lesion analysis is as follows:

[0166] The system automatically identifies the main trunk and major branches of the coronary arteries in scanned images using deep learning models or traditional image processing algorithms, and marks the identified main trunk and major branches of the coronary arteries as suspicious lesion areas.

[0167] Plaques in areas of suspicion are classified into plaque types, including calcified plaques, non-calcified plaques, and mixed plaques.

[0168] Next, the degree of stenosis in the suspected lesion area is assessed, including measuring the ratio of the inner diameter or cross-sectional area of ​​the stenotic vessel to the proximal normal vessel using the diameter method or area method, and calculating the stenosis impact assessment score in conjunction with the contrast agent filling status shown by angiography.

[0169] Based on the stenosis impact assessment score, it is determined whether the stenosis leads to hemodynamic changes;

[0170] High-risk features are identified based on plaque classification and stenosis, and a visual report is generated from the identified high-risk features.

[0171] In this embodiment, when the ratio of the inner diameter of the stenotic vessel to the proximal normal vessel exceeds a set inner diameter ratio threshold, the difference between the ratio of the inner diameter of the stenotic vessel to the proximal normal vessel and the inner diameter ratio of the set inner diameter ratio threshold is divided by the set inner diameter ratio threshold to obtain a key ratio difference. This difference is then combined with the contrast agent filling status shown by angiography to calculate the stenosis impact assessment score. The set inner diameter ratio threshold is determined based on a large amount of clinical data and imaging studies. The inner diameter ratio difference is the difference between the ratio of the inner diameter of the stenotic vessel to the proximal normal vessel and the set inner diameter ratio threshold.

[0172] When the ratio of the cross-sectional area of ​​the stenotic vessel to the proximal normal vessel exceeds a set cross-sectional area ratio threshold, the difference between this ratio and the set cross-sectional area ratio threshold is divided by the set cross-sectional area ratio threshold to obtain the critical ratio difference. This difference is then combined with the contrast agent filling status shown on angiography to calculate the stenosis impact assessment score. The set cross-sectional area ratio threshold is determined based on a large amount of clinical data and imaging studies. The cross-sectional area ratio difference is the difference between the ratio of the cross-sectional area of ​​the stenotic vessel to the proximal normal vessel and the set cross-sectional area ratio threshold.

[0173] When the ratio of the inner diameter of the stenotic vessel to the proximal normal vessel exceeds a set inner diameter ratio threshold, and the ratio of the cross-sectional area of ​​the stenotic vessel to the proximal normal vessel exceeds a set cross-sectional area ratio threshold, the first ratio (inner diameter ratio difference divided by the set inner diameter ratio threshold) and the second ratio (cross-sectional area ratio difference divided by the set cross-sectional area ratio threshold) are calculated. The maximum value of the first and second ratios is then used as the critical ratio difference. Combined with the contrast agent filling status shown on angiography, the stenosis impact assessment score is calculated.

[0174] In this embodiment, the formula for calculating the narrowing impact assessment score is as follows:

[0175]

[0176] In the formula, Y represents the narrowing impact assessment score; Represented as the critical percentage difference; This represents the weighted contribution of the degree of vascular stenosis to the assessment of its hemodynamic changes, with values ​​ranging from [value range missing]. ; This represents the weighting of the contribution of contrast agent filling status to the analysis of hemodynamic changes, with a value range of [value missing]. ; This is represented as the total number of times the filling time is compared; This represents the number of times the filling time ratio exceeds a set time ratio threshold; This represents the filling time ratio when the i-th filling time ratio exceeds the set time ratio threshold; This is expressed as a set time ratio threshold; Represented as the average value of the quantification of fullness; This represents the difference between the current quantitative value of fullness and the average value of fullness when the i-th fullness time ratio exceeds the set time ratio threshold; e represents the base of the natural logarithm, with a value of 2.7; ln represents the natural logarithm.

[0177] The weights assigned to the degree of vascular stenosis and the contrast agent filling status are obtained by solving the matrix constructed after pairwise comparison and scoring using the analytic hierarchy process (AHP). The total number of filling time comparisons refers to the total number of comparisons of the filling time of the contrast agent in the stenotic area and the proximal normal vessel. The specific degree of filling in each comparison is different (quantified by the average gray value of the region measured by image processing software on the selected region of interest (usually a region that can represent the filling status of the vessel, such as a specific length or range of the stenotic area and the proximal normal vessel).

[0178] Among them, the set time ratio threshold is a critical value determined based on a large amount of clinical data, experimental research, and experience, used to judge the degree of abnormality in the contrast agent filling time ratio between the stenotic site and the proximal normal vessel; the filling time ratio is the ratio of the time required for the stenotic site to fill to a specific degree to the time required for the contrast agent to fill to the same specific degree in the proximal normal vessel; the filling quantification average is obtained by adding the quantification values ​​of the corresponding filling degree for each filling time comparison and then averaging them; the filling degree quantification value is the average gray value of the region obtained by measuring the selected region of interest using image processing software.

[0179] In this embodiment, when the impact assessment score obtained after normalizing the stenosis impact assessment coefficient exceeds the set assessment threshold, it is determined that the stenosis causes hemodynamic changes; if the impact assessment score does not exceed the set assessment threshold, it is determined that the stenosis does not cause hemodynamic changes. The set assessment threshold is a pre-set value used to determine whether the stenosis causes hemodynamic changes, for example, 0.5.

[0180] The beneficial effects of the above technical solution are: by quantifying the impact of stenosis on hemodynamics to obtain a specific score, it can effectively reflect the severity of stenosis and its obstruction of blood flow, providing doctors with more intuitive and comparable assessment results, and benefiting subsequent treatment operations and achieving auxiliary diagnosis.

[0181] Specifically, image quality assessment is used as the screening criterion. Unqualified images are promptly optimized or rescanned to avoid misdiagnosis or missed diagnosis due to poor image quality. Optimization can specifically eliminate artifacts and enhance contrast, while rescanning obtains clearer data, ensuring that all images entering the lesion analysis stage meet diagnostic requirements. This lays a solid data foundation for subsequent accurate analysis. With the help of deep learning models or traditional image processing algorithms, the main coronary arteries and major branches are automatically identified, quickly locating suspicious lesion areas. Compared with manual identification, efficiency is improved by over 80%. At the same time, the algorithm can accurately delineate the vessel contour, reducing human judgment errors, providing accurate target areas for subsequent analysis, shortening diagnostic time, and classifying plaques and assessing the degree of stenosis in suspicious lesion areas, achieving multi-dimensional quantitative evaluation. Clear plaque classification distinguishes between calcified, non-calcified, and mixed plaques, helping physicians understand the nature of the lesion. Using diameter and area methods combined with contrast agent filling to assess the degree of stenosis and hemodynamic changes provides comprehensive information for clinical decision-making, such as whether interventional treatment is necessary, by identifying high-risk features and generating visual reports. Complex lesion information is presented in intuitive charts and annotated images, allowing physicians to quickly grasp key information. For example, clearly marking the location, degree of stenosis, and blood flow impact of high-risk plaques enables physicians to develop more efficient personalized treatment plans, while also facilitating doctor-patient communication, improving patient understanding of the condition, and optimizing the overall diagnosis and treatment process.

[0182] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0183] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A system for analyzing coronary artery lesions in elderly individuals during free breathing, characterized in that: The method comprises the following steps: First, the patient's information is collected from the database, and the parameters of the third-generation dual-source CT and high-pressure double-barrel injector are set according to the scanning requirements. When the patient is scanned, the scanning data is collected in real time, the real-time collected scanning data is reconstructed and processed, and the scanning image is generated. The generated scanning image is optimized, and the image quality of the optimized scanning image is evaluated. According to the quality evaluation result, the scanning image is analyzed for lesions; The generated scanning image is optimized, which comprises: According to the scanning range, the scanning image is manually or automatically cropped, and at the same time, the window width and window level are adjusted according to the characteristics of coronary artery imaging to highlight the contrast of blood vessel structure and surrounding soft tissue, fat and calcification; Adjusting the window width and window level according to the characteristics of coronary artery imaging comprises: Comparing and evaluating the contrast of the blood vessel structure in the current scanning image and the surrounding soft tissue, fat and calcification, and determining the preselected adjustment rule according to the evaluation result; According to whether the current patient has a history of coronary atherosclerosis, the corresponding screening condition is selected to match the coronary artery estimated CT range, the first window width and the first window level from the pre-set parameter mapping table; Based on the patient's real disease and disease development stage, the correlation level of coronary artery imaging quality is analyzed, and the real disease with a correlation level other than low correlation is regarded as a correlation analysis disease; If there is no correlation analysis disease in the history of the current patient, the first window width or the first window level is fine-tuned based on the preselected adjustment rule, and then the adjusted window width and the adjusted window level are outputted; If there is a correlation analysis disease in the history of the current patient, the corresponding disease estimated CT range is determined according to the disease development stage of the correlation analysis disease; According to the quality evaluation result, the scanning image is analyzed for lesions, which comprises: According to the image quality evaluation result, the scanning image that fails to meet the evaluation requirement is optimized or re-scanned; The scanning image with qualified image quality is analyzed for lesions; The lesion analysis comprises: Using a deep learning model or a traditional image processing algorithm to automatically identify the coronary artery trunk and main branches in the scanning image, and marking the identified coronary artery trunk and main branches as suspicious lesion areas; Classifying the suspicious lesion areas, and the plaque types include calcified plaque, non-calcified plaque and mixed plaque; Then, the stenosis degree of the suspicious lesion area is judged, which comprises measuring the ratio of the stenosis blood vessel diameter or cross-sectional area to the proximal normal blood vessel by using the diameter method or the area method, and calculating the stenosis impact evaluation score in combination with the contrast agent filling condition displayed by the angiography; According to the stenosis impact evaluation score, it is judged whether the stenosis leads to hemodynamic changes; According to the plaque classification and stenosis degree, high-risk features are identified, and the identified high-risk features are visualized and reported; When the ratio of the inner diameter of the stenosis to the normal blood vessel exceeds the set inner diameter ratio threshold, the ratio of the inner diameter of the stenosis to the normal blood vessel is divided by the inner diameter ratio difference of the set inner diameter ratio threshold to obtain the key ratio difference, and the stenosis impact assessment score is calculated in combination with the contrast agent filling shown by the angiography; wherein the set inner diameter ratio threshold is determined based on a large amount of clinical data, imaging research, etc.; the inner diameter ratio difference refers to the difference between the ratio of the inner diameter of the stenosis to the normal blood vessel and the set inner diameter ratio threshold; When the ratio of the cross-sectional area of the stenosis to the normal blood vessel exceeds the set cross-sectional area ratio threshold, the ratio of the cross-sectional area of the stenosis to the normal blood vessel is divided by the cross-sectional area ratio difference of the set cross-sectional area ratio threshold to obtain the key ratio difference, and the stenosis impact assessment score is calculated in combination with the contrast agent filling shown by the angiography; wherein the set cross-sectional area ratio threshold is determined based on a large amount of clinical data, imaging research, etc.; the cross-sectional area ratio difference refers to the difference between the ratio of the cross-sectional area of the stenosis to the normal blood vessel and the set cross-sectional area ratio threshold; When the ratio of the inner diameter of the stenosis to the normal blood vessel exceeds the set inner diameter ratio threshold, and the ratio of the cross-sectional area of the stenosis to the normal blood vessel exceeds the set cross-sectional area ratio threshold, the first ratio of the inner diameter ratio difference divided by the set inner diameter ratio threshold and the second ratio of the cross-sectional area ratio difference divided by the set cross-sectional area ratio threshold are calculated, and the maximum value of the first ratio and the second ratio is taken as the key ratio difference, and the stenosis impact assessment score is calculated in combination with the contrast agent filling shown by the angiography; The calculation formula of the stenosis impact assessment score is as follows: Y represents a narrow impact assessment score; Y represents a key proportion difference value; Y represents a contribution weight of a degree of vascular stenosis to assessment of hemodynamic changes, and the value range is ; Y represents a contribution weight of a contrast agent filling condition to analysis of hemodynamic changes, and the value range is ; Y represents a total number of filling time comparisons; Y represents a number of times that a filling time ratio exceeds a set time ratio threshold value; Y represents a corresponding filling time ratio when the i-th filling time ratio exceeds the set time ratio threshold value; Y represents a set time ratio threshold value; Y represents a filling quantification average value; Y represents a difference between a current filling degree quantification value and the filling quantification average value when the i-th filling time ratio exceeds the set time ratio threshold value; e represents a base number of a natural logarithm, and the value is 2.7; ln represents a natural logarithm.

2. The geriatric free-breathing coronary scan image lesion analysis system of claim 1, wherein, The real-time collected scanning data is reconstructed and scanning images are generated, including: After the scanning is completed, the projection data, synchronization data and device operation data collected by the CT device in real time are automatically transmitted to the image reconstruction workstation; The image reconstruction workstation processes the received data, including identifying each R-wave peak in the ECG signal in the data, dividing the heart into 10 equal phases, matching the respiratory signal with the timestamp of the projection data, and identifying the respiratory phase; After the data processing is completed, the image reconstruction workstation selects a reconstruction algorithm, including a conventional reconstruction and an ECG-gated reconstruction; After the reconstruction algorithm selection is completed, the reconstruction parameters are set, including the slice thickness, the slice interval, the field of view and the reconstruction matrix; The image reconstruction workstation performs three-dimensional reconstruction on the projection data according to the reconstruction algorithm and the reconstruction parameters to generate a plurality of cross-sectional original images, and the staff selects the image at the end of diastole as the main diagnostic scanning image through the dynamic browsing function.

3. The geriatric free-breathing coronary scan image lesion analysis system of claim 2, wherein, The patient's information is collected from the database, including: Before collecting the patient's information, the identity of the staff is authenticated, and after the authentication is passed, the patient's information is collected by logging into the database; The patient information includes basic information and clinical information; The basic information is patient name, department, doctor and patient identification; the clinical information includes examination application form, medical history and laboratory examination results; According to the collected patient information, the staff and the patient carry out on-site checking; If the checking is correct, the patient prepares for scanning; if the checking is incorrect, the staff adjusts the missing or abnormal information.

4. The geriatric free-breathing coronary scan image lesion analysis system of claim 3, wherein, According to the scanning requirements, the parameters of the third-generation dual-source CT and the high-pressure double-barrel injector are set, including: The scanning requirements are retrieved from the rule base, and the scanning type is confirmed according to the scanning requirements and patient information, including coronary artery enhancement scanning or calcification integral scanning; After the scanning type is confirmed, the parameters of the third-generation dual-source CT and the high-pressure double-barrel injector are set in turn; The parameter setting steps of the third-generation dual-source CT are as follows: Use retrospective ECG-gated helical scanning or prospective ECG-triggered axial scanning, and enable respiratory motion compensation technology; The scanning range is from the tracheal carina level to 1 cm below the diaphragm, covering the coronary artery origin to the distal branch; The ray parameters are set, including tube voltage and tube current, wherein the tube voltage is 100 kV or 120 kV, the tube voltage is 100 kV for patients with body weight < 70 kg or the elderly, and the tube voltage is 120 kV for patients with body weight ≥ 70 kg or severe calcification; the tube current adopts automatic tube current modulation technology, which is automatically adjusted according to the patient's body type and scanning site, and the tube current of the patient with low BMI is manually reduced; The pitch and layer thickness are set, wherein the pitch is calculated according to the heart rate, the pitch is set to 0.2-0.3, the layer thickness is 0.5-0.75 mm, and the interlayer spacing is 0.5 mm; The parameter setting steps of the high-pressure double-barrel injector are as follows: Select isotonic or hypotonic non-ionic contrast agent, the concentration is 350-400 mgI / mL, and the contrast agent dosage is calculated according to the patient's body weight; The injection rate is 5.0-6.0 mL / s, the elderly reduce to 4.5-5.0 mL / s, the physiological saline flushing dose is 30-50 mL, and the injection pressure is ≤300 PSI; Finally, the parameters of the third-generation dual-source CT and the high-pressure double-barrel injector are set.

5. The geriatric free-breathing coronary scan image lesion analysis system of claim 4, wherein, When scanning the patient, real-time scanning data is collected, including: First, guide the patient into the scanning room and connect the life monitoring equipment; Start the third-generation dual-source CT for positioning image scanning, obtain the patient's chest projection image, adjust the scanning range according to the positioning image, and check again whether the CT parameter setting meets the scanning requirements; After meeting the requirements, formal scanning and contrast agent injection are carried out, wherein the injection program is started according to the preset parameters, the contrast agent is injected first, and then the physiological saline flushing is automatically switched, after the contrast agent injection starts, the staff triggers the scanning instruction on the CT console, and the ECG-gated synchronization and respiratory motion compensation are started synchronously; The ECG-gated synchronization is for retrospective ECG-gated helical scanning, which continuously records the ECG signal and matches the scanning data with the heart movement phase; the respiratory motion compensation is to enable the respiratory gating or motion tracking technology of the CT equipment, which records the respiratory amplitude in real time; The CT tube and the detector rotate around the chest of the patient to collect projection data layer by layer, and record synchronous data including an electrocardiogram signal, a breathing signal, a gantry selection angle, a radiation dose and a scanning time; Finally, the projection data, the synchronous data and the equipment operation data are obtained, and the real-time acquisition of the scanning data is completed.

6. The geriatric free-breathing coronary scan image lesion analysis system of claim 5, wherein, The generated scanning image is subjected to image optimization processing, and the image optimization processing further includes: The scanning image is corrected and denoised, the correction is phase alignment of image dislocation caused by breathing motion by using the recorded breathing signal during scanning, and multi-phase image fusion or interpolation processing is performed on the blood vessel blur area caused by heart beating based on electrocardiogram gating data; the denoising is to reduce the granular artifacts caused by the radiation dose and equipment noise during scanning by using non-sharpening filtering or anisotropic diffusion algorithm, while the blood vessel edge details are retained; The scanning image is subjected to image enhancement after the correction and denoising, including using a gradient-based edge enhancement algorithm or a blood vessel tensor filter to highlight the coronary artery lumen boundary, expanding the gray difference between the blood vessels and the plaques through histogram adjustment, and compensating for the partial volume effect caused by the layer thickness through adjacent layer image interpolation or three-dimensional convolution processing; Finally, the scanning image subjected to the image enhancement is reorganized into a coronal plane, a sagittal plane and a curved surface image at any angle, is unfolded into a planar image along the running of the coronary artery, and a two-dimensional MIP image of a blood vessel tree is generated to highlight the contrast agent-filled lumen structure; Finally, the optimization processing of the scanning image is completed.

7. The geriatric free-breathing coronary scan image lesion analysis system of claim 6, wherein, According to the characteristics of coronary artery imaging, the window width and the window level are adjusted, and the adjusting further includes: obtaining a disease prediction CT range related to analysis of a disease, and a prediction CT intermediate range of the coronary artery prediction CT range; if there is only a single prediction CT intermediate range, determining a second window width and a second window level according to the current prediction CT intermediate range; if there are multiple prediction CT intermediate ranges, extracting the upper limit and the lower limit of each prediction CT intermediate range, sorting the upper limit and the lower limit respectively, taking the intermediate value as a comprehensive CT intermediate range, and determining the second window width and the second window level according to the comprehensive CT intermediate range; comprehensively analyzing the first window width and the first window level, and the second window width and the second window level by introducing disease relevance to obtain a new window width and a new window level; adopting a preselected adjustment rule to fine-tune the new window width or the new window level, and then outputting the fine-tuned new window width or new window level as an adjusted window width and an adjusted window level.

8. The geriatric free-breathing coronary scan image lesion analysis system of claim 1, wherein, The image quality of the scanning image subjected to the optimization processing is evaluated, including: performing preliminary quality screening on the optimized scanning image, including detecting motion artifacts, radiation noise and equipment artifacts of the optimized scanning image, and performing contrast evaluation after the detection is completed; performing blood vessel structure display evaluation after the preliminary quality screening is completed, including branch display and edge sharpness evaluation of the blood vessels; performing breathing motion influence and heart beating influence evaluation after the blood vessel structure display evaluation is completed; obtaining the preliminary quality screening, the blood vessel structure display evaluation, the breathing motion influence and the heart beating influence evaluation, and performing quantitative evaluation after the obtaining; dividing the optimized scanning image into quality grades according to the quantitative evaluation result, including qualified quality, slight abnormality and severe abnormality.

Images