Search

CN-121998962-A - Coronary bypass blood vessel patency OCT-FFR fusion image evaluation method and system

CN121998962ACN 121998962 ACN121998962 ACN 121998962ACN-121998962-A

Abstract

The invention relates to the field of medical image processing, in particular to a coronary bypass blood vessel patency OCT-FFR fusion image evaluation method and system, which comprises the steps of firstly acquiring OCT images and FFR data of coronary bypass blood vessels of a patient, and performing multi-scale wavelet transformation on the OCT images to obtain characteristic images after wavelet transformation; the method comprises the steps of identifying an interesting region OCT-ROI in an OCT image based on a characteristic image, identifying a functional abnormality region FFR-ROI based on FFR data, overcoming the limitation of single-mode evaluation by fusing OCT morphological information and FFR functional information, and providing an objective basis for clinical treatment decision based on a risk classification mechanism of multidimensional similarity.

Inventors

  • HOU MING

Assignees

  • 川北医学院附属医院

Dates

Publication Date
20260508
Application Date
20260205

Claims (10)

  1. 1. The coronary bypass blood vessel patency OCT-FFR fusion image evaluation method is characterized by comprising the following steps: Acquiring an Optical Coherence Tomography (OCT) image and Fractional Flow Reserve (FFR) data of a coronary bypass vessel of a patient; Performing multi-scale wavelet transformation on the OCT image to obtain a characteristic image after wavelet transformation; Identifying a region of interest OCT-ROI in the OCT image based on the wavelet transformed feature image; Identifying a dysfunction region FFR-ROI based on the FFR data; Establishing a spatial mapping relationship between the OCT-ROI and the FFR-ROI; Calculating a multi-dimensional similarity between the OCT-ROI and the FFR-ROI; Evaluating the patency of the coronary bypass vessel of the patient according to the multidimensional similarity; The identifying a region of interest OCT-ROI in an OCT image based on the wavelet transformed feature image comprises: constructing a multidimensional feature description system comprising a morphological feature set, a texture feature set and a color feature set; Wherein the color feature set is established based on the variance and average of three components of hue, saturation and brightness of the HSI color space; Based on the multidimensional feature description system, the OCT image is segmented through a fuzzy C-means FCM clustering algorithm, and an interesting region OCT-ROI in the OCT image is obtained.
  2. 2. The method of claim 1, wherein the acquiring optical coherence tomography OCT image and fractional flow reserve FFR data of the patient's coronary bypass vessel comprises: Synchronously measuring multi-dimensional blood vessel images by adopting an intravascular optical coherence tomography catheter, and continuously acquiring the OCT images from a proximal blood vessel inlet; the FFR data of the coronary bypass vessel are detected by a pressure measuring guidewire.
  3. 3. The method of evaluating coronary bypass vessel patency OCT-FFR fusion image according to claim 1, wherein said performing a multi-scale wavelet transform process on the OCT image comprises: selecting a coif5 wavelet basis to carry out wavelet transformation on the OCT image to obtain wavelet coefficients of a plurality of decomposition levels; Thresholding the wavelet coefficients of the plurality of decomposition levels, wherein the threshold of the low-frequency image is set to 80% of the total number of pixels of the image, and the threshold of the high-frequency image is set to 90% of the total number of pixels of the image; dividing the low-frequency image and the high-frequency image into a plurality of segments, and calculating the average value of each segment; And obtaining the characteristic image after wavelet transformation based on the average value of each segment.
  4. 4. The coronary bypass vessel patency OCT-FFR fusion image evaluation method of claim 1, wherein a spatial mapping relationship between the OCT-ROI and the FFR-ROI is established; Calculating a multi-dimensional similarity between the OCT-ROI and the FFR-ROI; and evaluating the patency of the coronary bypass blood vessel of the patient according to the multidimensional similarity.
  5. 5. The coronary bypass vessel patency OCT-FFR fusion image evaluation method of claim 1, wherein the identifying a functional abnormality region FFR-ROI based on the FFR data comprises: preprocessing the FFR data, including signal smoothing, baseline correction, outlier filtering and data normalization; extracting pressure gradient characteristics, time sequence characteristics, waveform morphological characteristics and statistical characteristics of the FFR data; Based on the extracted features, the FFR data is segmented through a fuzzy C-means FCM clustering algorithm, so that a functional abnormality region FFR-ROI is obtained; Wherein, the fractional flow reserve value corresponding to the FFR-ROI of the abnormal function region is less than 0.8.
  6. 6. The coronary bypass vessel patency OCT-FFR fusion image evaluation method of claim 4, wherein the establishing a spatial mapping relationship between the OCT-ROI and the FFR-ROI comprises: establishing a unified reference coordinate system with a long axis of a blood vessel as a Z axis; Transforming OCT-ROI coordinates and FFR-ROI coordinates to the unified reference coordinate system, respectively; Identifying anatomical landmark points in the OCT-ROI and FFR-ROI, including vessel branches and calcification points; performing a preliminary registration between the OCT-ROI and the FFR-ROI based on the anatomical landmark points; and carrying out fine registration on the local region by adopting a non-rigid deformation model to obtain a spatial mapping relation between the OCT-ROI and the FFR-ROI.
  7. 7. The coronary bypass vessel patency OCT-FFR fusion image evaluation method of claim 4, wherein the calculating a multi-dimensional similarity between the OCT-ROI and the FFR-ROI comprises: calculating pixel-level similarity, including pixel value difference, region overlapping degree, edge consistency and intensity distribution similarity; calculating feature level similarity, including shape similarity, texture similarity, direction similarity and scale similarity; calculating semantic level similarity, including lesion type consistency, severity relevance, risk assessment consistency and treatment indication relevance; And weighting and fusing the pixel-level similarity, the feature-level similarity and the semantic-level similarity by adopting an adaptive weight mechanism to obtain the multidimensional similarity between the OCT-ROI and the FFR-ROI.
  8. 8. The coronary bypass vessel patency OCT-FFR fusion image evaluation method of claim 7, wherein the adaptive weighting mechanism comprises: Setting an initial weight based on expert experience; dynamically adjusting the weight according to the image quality, the feature significance and the consistency evaluation; continuously optimizing the weight through historical data analysis, expert feedback and pattern recognition; the weights are ensured to meet the normalization constraint that the sum of all weights is equal to 1.
  9. 9. The method for evaluating coronary bypass vessel patency OCT-FFR fusion image according to claim 8, wherein said evaluating patency of the patient coronary bypass vessel according to the multi-dimensional similarity comprises: Setting a similarity threshold, including a low risk threshold and a high risk threshold; when the multidimensional similarity is lower than the low risk threshold, judging that the bypass blood vessel is good in smoothness; when the multi-dimensional similarity is between the low risk threshold and the high risk threshold, judging that the bypass blood vessel has potential risk, and needing periodic follow-up; and when the multi-dimensional similarity is higher than the high risk threshold and the OCT-ROI and the FFR-ROI are completely matched in space, judging that the bypass blood vessel has occlusive lesions, and adjusting an antiplatelet drug scheme to strengthen antithrombotic treatment.
  10. 10. The coronary bypass vessel patency OCT-FFR fusion image evaluation system implementing the method of any one of claims 1-9, comprising: The data acquisition module is used for acquiring an OCT image and Fractional Flow Reserve (FFR) data of a coronary bypass blood vessel of a patient; the wavelet transformation processing module is used for performing multi-scale wavelet transformation processing on the OCT image to obtain a characteristic image after wavelet transformation; an OCT region identification module for identifying a region of interest OCT-ROI in the OCT image based on the wavelet transformed feature image; an FFR region identification module for identifying a functional abnormality region FFR-ROI based on the FFR data; a spatial mapping module for establishing a spatial mapping relationship between the OCT-ROI and the FFR-ROI; a similarity calculation module for calculating a multidimensional similarity between the OCT-ROI and the FFR-ROI; and the patency evaluation module is used for evaluating the patency of the coronary bypass blood vessel of the patient according to the multidimensional similarity.

Description

Coronary bypass blood vessel patency OCT-FFR fusion image evaluation method and system Technical Field The invention relates to the field of medical image processing, in particular to a coronary bypass blood vessel patency OCT-FFR fusion image evaluation method and system, and particularly relates to a method and system for coronary bypass blood vessel patency evaluation by combining Optical Coherence Tomography (OCT) and Fractional Flow Reserve (FFR) bimodal data. Background Currently, evaluation of bypass vascular patency relies mainly on two techniques, optical Coherence Tomography (OCT) and Fractional Flow Reserve (FFR) measurement. OCT techniques can provide high resolution vascular morphology information that can clearly show vessel wall structure and plaque characteristics, but cannot directly assess vessel function status. FFR measurements can then provide quantitative parameters of the functional state of the blood vessel, reflecting the hemodynamic properties of the blood vessel, but lack a description of the microstructure of the blood vessel. Both techniques have advantages, but are limited when used alone, and it is difficult to comprehensively evaluate the patency of the bypass blood vessel. In the prior art, OCT and FFR are usually used independently, and a doctor needs to comprehensively judge the two examination results by experience, so that an objective and quantitative fusion evaluation method is lacked. In addition, since OCT provides morphological information and FFR provides functional information, the two information modalities are different, and it is difficult to directly fuse analysis, which also limits the accuracy and reliability of the assessment. Therefore, how to effectively integrate information of two different modes of OCT and FFR, and realize comprehensive and accurate assessment of the patency of the coronary bypass blood vessel, becomes a technical problem to be solved urgently at present. Disclosure of Invention The invention aims to provide a coronary bypass blood vessel patency OCT-FFR fusion image evaluation method and system, which aim to overcome the limitation of single-mode evaluation in the prior art, and realize comprehensive and accurate evaluation of coronary bypass blood vessel patency by fusing OCT morphological information and FFR functional information. The invention provides a coronary bypass blood vessel patency OCT-FFR fusion image evaluation method, which comprises the following steps: Acquiring an Optical Coherence Tomography (OCT) image and Fractional Flow Reserve (FFR) data of a coronary bypass vessel of a patient; Performing multi-scale wavelet transformation on the OCT image to obtain a characteristic image after wavelet transformation; Identifying a region of interest OCT-ROI in the OCT image based on the wavelet transformed feature image; Identifying a dysfunction region FFR-ROI based on the FFR data; The identifying a region of interest OCT-ROI in an OCT image based on the wavelet transformed feature image comprises: constructing a multidimensional feature description system comprising a morphological feature set, a texture feature set and a color feature set; wherein the color feature set is established based on the variance and average of three components of hue, saturation and brightness of the HIS color space; Based on the multidimensional feature description system, the OCT image is segmented through a fuzzy C-means FCM clustering algorithm, and an interesting region OCT-ROI in the OCT image is obtained. Preferably, the acquiring the OCT image and fractional flow reserve FFR data of the patient coronary bypass vessel comprises: Synchronously measuring multi-dimensional blood vessel images by adopting an intravascular optical coherence tomography catheter, and continuously acquiring the OCT images from a proximal blood vessel inlet; the FFR data of the coronary bypass vessel are detected by a pressure measuring guidewire. Preferably, the performing the multi-scale wavelet transform processing on the OCT image includes: selecting a coif5 wavelet basis to carry out wavelet transformation on the OCT image to obtain wavelet coefficients of a plurality of decomposition levels; Thresholding the wavelet coefficients of the plurality of decomposition levels, wherein the threshold of the low-frequency image is set to 80% of the total number of pixels of the image, and the threshold of the high-frequency image is set to 90% of the total number of pixels of the image; dividing the low-frequency image and the high-frequency image into a plurality of segments, and calculating the average value of each segment; And obtaining the characteristic image after wavelet transformation based on the average value of each segment. Preferably, a spatial mapping relation between the OCT-ROI and the FFR-ROI is established; Calculating a multi-dimensional similarity between the OCT-ROI and the FFR-ROI; and evaluating the patency of the coronary bypass blood vessel o