Automated Lesion Segmentation and Classification via DCE-MRI Kinetic Texture Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current breast cancer screening methods, particularly x-ray mammography, are less effective in detecting triple-negative breast cancer and face challenges in accurately differentiating between benign and malignant lesions due to high inter-observer variability and limited sensitivity, especially in dense breast tissue and younger patients.
Innovation Solution
A computer-aided diagnosis system utilizing DCE-MRI images processes kinetic texture features to classify lesions as malignant or benign and specifically differentiate between triple-negative and non-triple-negative breast cancers, employing an Expectation Maximization-driven Active Contour scheme for lesion segmentation and a support vector machine for classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If x-ray mammography is used for breast cancer screening, then the screening process is simple and widely available, but the detection sensitivity is reduced especially for triple-negative breast cancer and dense breast tissue
Solution Approach 1:
The patent uses DCE-MRI as an intermediary imaging modality between conventional mammography and pathological diagnosis. The dynamic contrast enhancement patterns provide intermediate information about lesion vascularity and perfusion characteristics, enabling better differentiation of triple-negative breast cancer while maintaining a structured diagnostic workflow
Solution Approach 2:
The patent transforms the diagnostic approach by changing from static anatomical imaging parameters (mammography) to dynamic functional parameters (contrast enhancement kinetics). By analyzing temporal changes in signal intensity and texture features during contrast agent passage, the system achieves improved detection sensitivity for aggressive breast cancer subtypes
2Measurement precision
If qualitative radiologic descriptors are used for TN breast cancer classification, then the classification process is simple, but the measurement precision and objectivity are reduced due to high inter-observer variability
Solution Approach 1:
The patent replaces the manual, subjective radiologist assessment (mechanical/visual system) with an automated computer-aided diagnosis system. The system uses algorithms to objectively quantify kinetic texture features from DCE-MRI sequences, eliminating inter-observer variability and providing consistent, reproducible classification of triple-negative breast cancer lesions
Solution Approach 2:
The patent creates a computational model that copies and analyzes the temporal evolution of contrast enhancement patterns. By generating synthetic kinetic texture features that mirror the actual imaging data, the system enables automated classification while preserving the essential diagnostic information
3Reliability
If DCE-MRI is used to detect TN breast cancer, then the detection sensitivity is improved, but the device complexity and cost increase
Solution Approach 1:
The patent segments the DCE-MRI diagnostic process into distinct analytical components: temporal signal intensity analysis, kinetic curve fitting, and texture feature extraction. This segmentation allows the system to process complex DCE-MRI data through manageable stages, reducing computational complexity while maintaining high detection sensitivity for triple-negative breast cancer
Data Source
AI summary
A method and apparatus for classifying possibly malignant lesions from sets of DCE-MRI images includes receiving a set of MRI slice images obtained at respectively different times, where each slice image includes voxels representative of at least one region of interest (ROI). The images are processed to determine the boundaries of the ROIs and the voxels within the identified boundaries in corresponding regions of the images from each time period are processed to extract kinetic texture features. The kinetic texture features are then used in a classification process which classifies the ROIs as malignant or benign. The malignant lesions are further classified to separate TN lesions from non-TN lesions.