Learning Support Device for Medical Region-of-Interest Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the medical field, manually discriminating useful training data from massive datasets for region-of-interest identification in medical images is inefficient and unreasonable.
Innovation Solution
A learning support device comprising a storage unit, acquisition unit, registration unit, and learning unit that analyzes interpretation reports to acquire and register images and names of regions of interest, generating a discrimination model using machine learning for accurate region-of-interest identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual discrimination is used to select training data from massive datasets, then data quality can be ensured, but work efficiency is extremely low and time-consuming
Solution Approach 1:
The system performs self-service by automatically analyzing interpretation reports to extract training data without human intervention. The acquisition unit automatically parses medical images and text information, extracts region-of-interest data, and registers training data pairs, enabling the system to select and prepare training data autonomously while maintaining quality standards.
Solution Approach 2:
The patent replaces the mechanical manual discrimination process with an automated information processing system. The acquisition unit uses computer-based algorithms to analyze interpretation reports, extract relevant features, and generate training data, substituting human manual work with automated computational processes that significantly improve efficiency while maintaining data quality.
2Quantity of substance
If massive data is processed manually, then comprehensive training data can be acquired, but the process is unreasonable and inefficient
Solution Approach 1:
The system autonomously processes massive datasets by automatically analyzing interpretation reports, extracting region-of-interest information, and generating training data pairs without human intervention. The acquisition unit continuously processes data from the database, enabling scalable processing of large volumes of medical imaging data while maintaining high efficiency.
Solution Approach 2:
The patent replaces manual data processing with automated computational systems that can efficiently handle massive datasets. The acquisition unit uses computer-based algorithms to parse, analyze, and extract training data from interpretation reports at scale, enabling the processing of large volumes of data that would be impractical to handle manually.
3Measurement precision
If deep learning is used for region-of-interest identification, then accuracy can be improved, but massive high-quality training data is required which is difficult to acquire manually
Solution Approach 1:
The system autonomously acquires training data by automatically analyzing interpretation reports and extracting region-of-interest information. The acquisition unit parses medical images and text, identifies relevant features, and generates training data pairs without human intervention, providing the massive high-quality training data required for deep learning model development.
Solution Approach 2:
The patent replaces the complex manual process of training data acquisition with an automated information processing system. The acquisition unit uses computer-based algorithms to efficiently extract and structure training data from interpretation reports, significantly reducing the complexity of data acquisition while providing sufficient high-quality data for deep learning applications.
Data Source
AI summary
A learning support device 18 includes an acquisition unit 26, a registration unit 27, a storage device 28, a learning unit 29, and a controller 31. The acquisition unit 26 acquires an image of a region of interest and a name of the region of interest by analyzing an interpretation report 23. The registration unit 27 registers training data consisting of the image of the region of interest and the name of the region of interest acquired by the acquisition unit 26 in the storage device 28. The learning unit 29 performs learning for generating a discrimination model 34, which outputs the image of the region of interest and the name of the region of interest with respect to an input of an inspection image 22 of the interpretation report 23, using a plurality of pieces of training data 33 registered in the storage device 28.