Medical Image Classification Reliability via Multi-Position Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical image processing technologies face instability in classification results due to body motion and changes in imaging position, leading to unreliable highlighted displays of lesions in medical images.
Innovation Solution
A medical image processing apparatus that includes an image acquisition unit, a classifier, a reliability calculation unit, and a confirmation unit to classify medical images into classes based on feature amounts, calculate reliability, and confirm classification results only when reliability meets specific conditions, such as threshold values or consistent results across multiple images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If classification is performed on medical images, then lesion identification is achieved, but classification stability deteriorates due to body motion and imaging position changes
Solution Approach 1:
The system performs preliminary actions by acquiring multiple images at different positions before final classification. The image acquisition unit captures a plurality of images at different imaging positions, and the classification unit uses these pre-acquired images to determine classification results, thereby compensating for subsequent body motion or position changes.
Solution Approach 2:
The system implements feedback mechanisms where the classification result is fed back to the image acquisition unit. When body motion or position change is detected, the system re-acquires images based on the classified region and adjusts imaging parameters accordingly, creating a closed-loop system that maintains classification stability.
2Reliability
If multiple images are acquired at different positions, then classification reliability is improved, but imaging time and data processing complexity increase
Solution Approach 1:
The system applies local quality by acquiring images selectively at different positions based on the classified region. Instead of uniformly acquiring multiple images throughout the entire medical image, the system focuses on specific regions where motion or position changes are detected, thereby reducing overall imaging time while maintaining reliability.
Solution Approach 2:
The system uses partial action by acquiring only the necessary number of images at different positions required for reliable classification. The image acquisition unit acquires a plurality of images at different imaging positions, but only processes and analyzes those necessary for the classified region, avoiding excessive imaging and processing.
3Measurement precision
If classification results are confirmed based on reliability thresholds, then false positives are reduced, but processing steps and system complexity increase
Solution Approach 1:
The system implements self-service by having the classification unit automatically evaluate its own results against reliability thresholds. The classification unit determines whether to output a classification result based on internally calculated reliability metrics, eliminating the need for external verification steps and reducing overall system complexity.
Data Source
AI summary
A medical image processing apparatus includes an image acquisition unit that acquires a medical image, a classifier that classifies the medical image or a region of interest included in the medical image into any of two or more classes based on a feature amount obtained from the medical image, a reliability calculation unit that calculates reliability of a classification result of the medical image or the region of interest from the classifier, and a confirmation unit that confirms the classification result of the medical image or the region of interest based on the calculated reliability.