Morphological Abnormality Detection Feedback for Tissue Specimens
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing drug discovery methods suffer from inaccurate detection of morphological abnormalities in tissue specimens, leading to overdetection or underdetection, which affects the evaluation of candidate substances.
Innovation Solution
A drug discovery support device and method that utilize a processor to detect, derive, and present cause identification reference information for overdetected or underdetected morphological abnormalities, using machine learning models and determination thresholds to enhance detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic image analysis is used to detect morphological abnormalities, then productivity is improved, but measurement precision deteriorates due to overdetetection or underdetection
Solution Approach 1:
The system generates determination reference information based on detection results and feeds it back to help identify causes of overdetetection or underdetection. This feedback mechanism allows continuous improvement of detection accuracy while maintaining automatic analysis efficiency.
Solution Approach 2:
The patent introduces determination reference information as an intermediary between the automatic detection system and final accuracy assessment. This intermediary element helps bridge the gap between automated speed and precision by providing reference data for evaluating detection quality.
2Measurement precision
If detection sensitivity is increased to reduce underdetection, then measurement precision improves, but overdetection increases
Solution Approach 1:
The system uses determination reference information to provide feedback on detection performance, enabling identification of overdetection causes without requiring increased sensitivity. This allows maintaining optimal sensitivity levels while improving overall accuracy through cause analysis.
Solution Approach 2:
The patent changes the parameter being measured from raw detection counts to determination reference information that indicates detection quality. This parameter transformation allows assessing detection accuracy without directly increasing detection sensitivity, thereby avoiding overdetection.
3Measurement precision
If expert review is performed to improve detection accuracy, then measurement precision improves, but productivity deteriorates
Solution Approach 1:
The system extracts and analyzes determination reference information from detection results to automatically identify causes of errors. This extraction approach captures the essential quality assessment function without requiring full expert review, maintaining productivity while improving precision.
Solution Approach 2:
The system performs self-assessment of detection quality by generating and analyzing determination reference information. This self-service mechanism provides automated accuracy evaluation, reducing the need for external expert intervention while maintaining high detection standards.
Data Source
AI summary
A processor is configured to obtain a specimen image of a subject; detect one or more estimated morphological abnormality portions; derive determination reference information from the one or more estimated morphological abnormality portions; determine, based on the determination reference information, whether the one or more estimated morphological abnormality portions have been overdetected and whether the one or more estimated morphological abnormality portions have been underdetected; in response to a determination being made that the one or more estimated morphological abnormality portions have been overdetected, present to a user first cause identification reference information contributing to identifying a cause of the one or more estimated morphological abnormality portions having been overdetected; and in response to a determination being made that the one or more estimated morphological abnormality portions have been underdetected, present to the user second cause identification reference information.