Identifier Selection for Abnormal Data in Medical Diagnosis
Solution Overview
Problem
Existing medical diagnosis assist systems face challenges in accurately evaluating the discrimination accuracy of identifiers for abnormal data due to differences in treatment data characteristics across medical institutions, leading to suboptimal performance for specific institutions.
Innovation Solution
An information processing device that customizes evaluation data to match the characteristics of each medical institution, using a hardware processor to calculate and select the most accurate identifier for discriminating abnormal data, thereby improving discrimination accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single identifier is selected based on discrimination accuracy calculated using general evaluation data, then the identifier with highest discrimination accuracy is applied, but the discrimination accuracy does not match the characteristics of individual medical institutions
Solution Approach 1:
The patent applies local quality by customizing evaluation data according to each medical institution's characteristics. Instead of using a single general evaluation dataset, the system creates institution-specific evaluation data that reflects local patient demographics, disease patterns, and treatment characteristics, thereby making the identifier selection process adapted to local conditions while maintaining high discrimination accuracy
Solution Approach 2:
The patent implements parameter changes by modifying the evaluation data parameters to match each medical institution's characteristics. The system changes the composition and properties of evaluation data based on institution-specific factors such as patient age distribution, disease types, and treatment protocols, enabling the identifier selection to be optimized for each institution rather than using fixed parameters
2Measurement precision
If evaluation data is customized for each medical institution, then discrimination accuracy matches institution characteristics, but the complexity of the evaluation process increases
Solution Approach 1:
The patent applies self-service by enabling medical institutions to automatically generate and customize their own evaluation data based on their treatment data. The system allows each institution to independently perform the evaluation data customization process using their own data resources, reducing the need for external intervention and simplifying the overall implementation despite the increased customization capability
Data Source
AI summary
An information processing device includes: a plurality of identifiers that discriminates whether abnormal data that is information indicating an abnormal value is included in treatment data and detects the abnormal data; and a hardware processor that: customizes evaluation data used when evaluating discrimination accuracy of the plurality of identifiers; calculates the discrimination accuracy of each of the plurality of identifiers using the evaluation data customized by the hardware processor; and selects one identifier out of the plurality of identifiers as an identifier used for discrimination based on a calculation result by the hardware processor.