Automated Contrast Phase Medical Image Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer-aided diagnosis (CAD) systems face inaccuracies due to the lack of automated mechanisms for selecting or excluding medical images with sufficient contrast material enhancement, leading to inefficient resource utilization and potential misdiagnosis.
Innovation Solution
A data processing system employing machine learning models for body part regression and segmentation to select a representative slice with sufficient contrast material enhancement, determining radiodensity metrics, and routing medical imaging studies to appropriate CAD systems for accurate processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all medical imaging slices are processed through CAD systems, then diagnostic accuracy is improved, but resource costs and processing time increase significantly
Solution Approach 1:
The patent segments the medical imaging study into a subset of representative slices for CAD processing. Instead of processing all slices, the system selects only those slices that are most likely to contain diagnostic information, thereby maintaining diagnostic accuracy while reducing resource consumption and processing time.
Solution Approach 2:
The patent extracts and processes only the necessary slices from the complete medical imaging study. By identifying and isolating the most relevant slices based on contrast enhancement metrics, the system eliminates unnecessary processing of non-diagnostic slices, improving efficiency without compromising diagnostic reliability.
2Productivity
If automated contrast phase classification is implemented, then resource utilization is improved, but system complexity increases
Solution Approach 1:
The patent implements self-service through automated contrast phase classification and representative slice selection. The system automatically evaluates contrast enhancement in medical imaging slices and identifies diagnostic slices without requiring manual radiologist intervention for slice selection, thereby improving resource utilization while the automation manages the complexity internally.
3Measurement precision
If manual slice selection is performed, then processing accuracy is maintained, but time consumption and labor costs increase
Solution Approach 1:
The patent replaces the manual mechanical process of radiologist slice selection with an automated computational system. The system uses contrast enhancement evaluation and regression analysis to automatically identify representative slices, maintaining selection accuracy while eliminating the time consumption and labor costs associated with manual review of all slices.
Data Source
AI summary
Mechanisms are provided for determining a measure of radiodensity of anatomical structures of interest and classifying medical imaging study data structures (studies) with regard to contrast phase. In some embodiments, this classification may be used to select/exclude slices for processing by other downstream computing systems. A subset of slices are selected from the study and, for each slice in the subset, a corresponding body part regression (BPR) score is determined. A linear regression on the BPR scores is performed and a representative slice is selected based on results of the linear regression. The representative slice is segmented and a statistical measure of a radiodensity metric for each segment in the representative slice is determined.