Clinical Information Processing Apparatus Using Likelihood Ratio Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing clinical information processing methods fail to accurately calculate the degree of similarity between a target patient's case and past cases due to insufficient weighting of clinical-information items, especially when the number of past cases is small, and do not appropriately account for the nonlinear changes in clinical-information item values or their standard ranges in medical diagnosis.
Innovation Solution
A clinical information processing apparatus and method that calculates a degree of similarity by determining weighting coefficients based on likelihood ratios for each clinical-information item, allowing for appropriate weighting even with a small number of past cases, and considers the likelihood of belonging to specific classifications to accurately reflect the relationship between clinical-information items and key items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the number of past cases is small, then the system can be implemented with limited data, but the accuracy of similarity calculation deteriorates
Solution Approach 1:
The patent transforms the similarity calculation from a direct comparison method to a likelihood ratio-based statistical method. By changing the calculation parameter from simple case matching to probability-based weighting using likelihood ratios, the system can achieve accurate similarity assessment even with limited past cases. The likelihood ratio for each clinical-information item is calculated based on the frequency distribution in available cases, allowing effective utilization of small datasets.
2Measurement precision
If weighting is applied based on case coincidence, then cases with high information matching are identified, but the inability to consider weighting on each symptom prevents appropriate evaluation when only slight coincidence exists
Solution Approach 1:
The patent applies local quality by assigning different weighting coefficients to different clinical-information items based on their individual likelihood ratios. Each symptom or clinical item is evaluated independently with its own weight, allowing the system to appropriately evaluate cases with slight coincidence in specific items while maintaining overall similarity assessment accuracy. This localized weighting approach avoids the need for complex global weighting mechanisms.
3Ease of manufacture
If the degree of influence is kept uniform for each clinical-information item, then the calculation method is simple, but it becomes impossible to accurately calculate similarity based on the characteristic of each item
Solution Approach 1:
The patent introduces dynamics by making the weighting coefficients adaptive rather than static. The likelihood ratio for each clinical-information item is dynamically calculated based on the frequency distribution in the available case data. This dynamic weighting mechanism automatically adjusts the importance of each item according to its actual discriminatory power in the dataset, achieving both calculation simplicity and high accuracy without requiring manual intervention.
4Ease of operation
If similarity is evaluated only based on difference between values, then the calculation is straightforward, but it is not appropriate when values should be judged against standard ranges or when changes are nonlinear
Solution Approach 1:
The patent applies preliminary action by pre-calculating the likelihood ratio for each clinical-information item based on the frequency distribution in the case database before actual similarity assessment. This preliminary statistical analysis establishes the appropriate weighting for each item, accounting for standard ranges and nonlinear relationships. When similarity evaluation is performed, these pre-established weights ensure that the assessment is both simple to execute and reliable in capturing clinically meaningful differences.
Data Source
AI summary
Likelihood ratio between a likelihood of belonging to one classification of a key item and a likelihood of belonging to other classification of the key item when a case belongs to each classification of a clinical-information item other than the key item is calculated, based on registration case information for calculating a likelihood ratio, for each classification of a key item. A weighting coefficient corresponding to each classification of the clinical-information item other than the key item for each classification of the key item is determined based on a target classification of a target clinical-information item and the calculated likelihood ratio. A degree of similarity is calculated for each registration case included in registration case information for calculating a degree of similarity by using weighting information corresponding to each classification of the key item and each classification of the clinical-information item other than the key item.