IVOCT Plaque Classification via SVM Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional imaging approaches for percutaneous coronary interventions (PCIs) lack effective visualization of significant calcium and lipid deposits, leading to uncertain treatment planning and increased complexity for cardiologists, as they do not provide sufficient information on the location, extent, and constituents of vascular lesions, especially during vessel remodeling.
Innovation Solution
The development of an automated system using machine learning classifiers, specifically support vector machine (SVM) classifiers, to classify intravascular plaque components in real-time from intravascular optical coherence tomography (IVOCT) images, extracting features such as optical, intensity, and spatial texture features to differentiate between fibrous, lipid, and calcified plaques, and generating annotated 3D visualizations for personalized treatment plans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional imaging approaches (intravascular ultrasound, x-ray angiography, fluoroscopy) are used to visualize vessel lumen, then basic structural information is obtained, but sufficient information on location, extent, and constituents of vascular lesions is not provided
Solution Approach 1:
The patent segments the complex task of plaque characterization into distinct classification categories (lipid, fibrous, calcified, mixed plaque types) using machine learning classifiers. This segmentation allows the system to provide specific information about lesion constituents without requiring a single complex imaging modality to detect all features simultaneously.
Solution Approach 2:
The patent introduces an intermediary computational layer (machine learning classification system) that processes IVOCT image data to extract and interpret plaque characteristics. This intermediary translates raw imaging data into meaningful clinical information about lesion constituents, bridging the gap between basic structural imaging and detailed compositional analysis.
2Measurement precision
If IVOCT is used to characterize coronary artery plaques with sufficient resolution and contrast, then discrimination between lipid and fibrous plaque constituents is achieved, but extensive specialized training is required and interpretation is time consuming
Solution Approach 1:
The patent implements a self-service system where the machine learning classification algorithm automatically analyzes IVOCT images and generates plaque characterizations without requiring physician interpretation. The system serves itself by processing images through trained classifiers that autonomously identify plaque types, eliminating the need for extensive specialized training and reducing interpretation time.
Solution Approach 2:
The patent replaces the manual mechanical process of physician image interpretation with an automated computational system. Machine learning classifiers substitute for the human expert's visual analysis, maintaining high measurement precision in plaque constituent discrimination while dramatically reducing the time required for interpretation.
3Area of stationary object
If a single IVOCT pullback is performed during PCI, then comprehensive vascular imaging is obtained, but over five hundred images are generated overloading the physician with data
Solution Approach 1:
The patent extracts only the essential information from the comprehensive IVOCT pullback data by using machine learning classifiers to identify and categorize plaque characteristics. Instead of presenting all five hundred images to the physician, the system extracts key findings (plaque type, location, extent) and presents them in a condensed, actionable format, reducing data complexity while preserving comprehensive vascular imaging coverage.
4Reliability
If conventional imaging approaches are used for PCI planning, then basic vessel lumen visualization is provided, but treatment planning accuracy regarding calcium and lipid deposits is compromised
Solution Approach 1:
The patent changes the analytical parameters of IVOCT imaging by applying machine learning classification algorithms that detect specific optical properties associated with different plaque constituents. This parameter transformation enables reliable identification of calcium and lipid deposits that conventional imaging cannot detect, thereby improving treatment plan reliability without losing information about deposit composition.
Data Source
AI summary
Methods and apparatus automatically classify intravascular plaque using features extracted from intravascular optical coherence tomography (IVOCT) imagery. One example apparatus includes an image acquisition circuit that accesses a set of IVOCT images, a pre-processing circuit that generates a blood vessel mask based on the IVOCT images, a feature extraction circuit that defines a three dimensional (3D) volume of interest centered on a location in a member of the set of IVOCT images, a classification circuit that generates a classification based on a probability that a voxel represents a type of plaque, and a visualization circuit that provides a visualization, substantially in real time, of a member of the set of IVOCT images and the classification, where the visualization includes a sector classification image, a labeled image, or a 3D visualization. A prognosis or treatment plan may be provided based on the visualization or the classification.