Two-Stage Disease Identification Model for Medical Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the medical field, there is a need for rapid recognition of patient severity, symptoms, and urgent treatment, which requires not only medical image identification but also effective presentation of identification results.
Innovation Solution
An information processing apparatus and method that utilize a first disease identification model for rapid recognition of body parts and disease presence, and a second model for detailed disease identification, facilitating quick and accurate patient assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single detailed disease identification model is used for comprehensive disease detection, then measurement precision is improved, but processing time increases and productivity decreases
Solution Approach 1:
The patent divides the disease identification task into two separate models: a first model for rapid preliminary identification and a second model for detailed analysis. This segmentation allows the system to achieve both speed and accuracy by processing different aspects of disease detection with specialized models rather than using a single comprehensive model.
Solution Approach 2:
The first disease identification model performs preliminary screening to quickly identify potential diseases and body portions. This preliminary action filters the data before it reaches the second model, reducing the processing load and enabling the detailed model to focus only on relevant cases, thus improving overall processing efficiency while maintaining accuracy.
2Measurement precision
If comprehensive disease identification is performed on all body portions, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The first disease identification model performs preliminary screening across all body portions to quickly identify potential diseases. This preliminary action enables the system to focus detailed analysis only on identified cases, reducing the overall time required while maintaining comprehensive detection coverage.
Solution Approach 2:
The two-stage modeling approach segments the time-consuming detailed analysis from the rapid initial screening. This allows comprehensive disease detection to be achieved without requiring all body portions to undergo lengthy detailed analysis simultaneously, thus reducing total processing time while maintaining detection completeness.
3Loss of information
If detailed disease information is provided for all cases, then information completeness is improved, but device complexity increases
Solution Approach 1:
The patent implements a segmented information provision strategy where the first model provides basic disease identification information quickly, and the second model provides detailed disease information only when needed. This segmentation reduces system complexity by avoiding the need for a single complex system to handle all levels of detail for all cases simultaneously.
Solution Approach 2:
The system applies partial detailed analysis only to cases identified by the first model rather than performing exhaustive detailed analysis on all cases. This partial action approach maintains information completeness for relevant cases while reducing the overall system complexity and resource requirements.
Data Source
AI summary
The present invention provides an information processing apparatus, an information processing method, an information processing program, and an information processing system that facilitates recognition of an identification result. An information processing apparatus has a first acquisition unit configured to acquire measurement data obtained by measurement of a physical condition of a patient, a second acquisition unit configured to acquire, using a first disease identification model generated by machine learning that uses teaching data labeled with first classifications of body portions included in measurement data obtained by measurement of a physical condition of a human and presence of a disease in the body portions, the first classification of a body portion in the measurement data acquired by the first acquisition unit and presence of a disease in the body portions, and a display unit configured to display the presence of the disease in the body portions that is acquired by the second acquisition unit in accordance with the first classification.