Information Processor for Disease Recurrence Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current clinical prediction systems for disease recurrence, such as liver cancer, are inadequate due to their reliance on non-theoretically optimized discriminant features and inability to handle qualitative data, limiting their accuracy and reliability.
Innovation Solution
An information processor that uses the discrete Bayes decision rule to predict the recurrence of diseases by calculating conditional and posterior probabilities based on feature subsets of both quantitative and qualitative data, allowing for the selection of optimal discriminant features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If scoring systems use non-numerical variables (qualitative data) for disease prediction, then the system can handle diverse clinical data types, but statistical information cannot be calculated and prediction reliability deteriorates
Solution Approach 1:
The patent transforms qualitative data into quantitative representations by assigning numerical codes to categorical values (e.g., encoding disease stages, treatment types, and outcomes as numbers). This parameter transformation enables statistical calculations while preserving the semantic meaning of qualitative clinical data, thereby maintaining both adaptability to diverse data types and reliability of predictions.
2Ease of manufacture
If discriminant features are determined through trial and error by medical doctors, then the system can be developed without extensive research funding, but the optimality of features is not theoretically assured and prediction accuracy deteriorates
Solution Approach 1:
The patent implements automated feature selection algorithms that independently identify optimal discriminant features from the data without requiring manual trial-and-error intervention by medical doctors. The system self-optimizes feature subsets based on statistical criteria and prediction performance, thereby maintaining ease of implementation while significantly improving prediction accuracy through theoretically grounded feature selection.
3Measurement precision
If molecular discriminant features are used to improve prediction accuracy, then the predictability of the scoring system is improved, but enormous research funding and long time are required for clinical trials and health insurance approval
Solution Approach 1:
The patent utilizes readily available, inexpensive clinical data that can be obtained through routine medical practice rather than expensive molecular biomarkers requiring lengthy clinical trials. By leveraging existing electronic health records and standard clinical measurements, the system achieves cost-effective and time-efficient prediction without the need for prolonged research and regulatory approval processes.
4Device complexity
If preliminary determined discriminant features are used in scoring systems, then the system structure is simplified, but if any data on discriminant features is defective, the scoring system cannot be used and reliability deteriorates
Solution Approach 1:
The patent implements dynamic feature selection that adapts to the specific characteristics of the input data rather than relying on fixed, pre-determined feature sets. The system automatically identifies and selects the most relevant features for each prediction case, adjusting the feature subset based on data quality and availability. This dynamic approach maintains relatively simple system structure while significantly improving reliability by avoiding failure when specific predetermined features are defective or unavailable.
Data Source
AI summary
An information processor can logically support prediction based on past statistical information even though the information contains qualitative or non-numerical data. The processor determines whether an input pattern corresponding to an input object (a determination target) belongs to a specific class among multiple classes, based on feature subsets of any combination of a plurality of features, each feature comprises multiple categories. The processor includes a storage storing the input pattern corresponding to the input object and samples corresponding to respective sample objects and a classification determiner determining whether the input pattern belongs to the specific class. The classification determiner calculates a first conditional probability and a second conditional probability based on the number of the samples belonging to each category of the respective features, the first conditional probability is a probability that the data of the input pattern belong to categories corresponding to the respective feature for the specific class, the second conditional probability is a probability that the data of the input pattern belong to categories corresponding to the respective features for a non-specific class which is a class other than the specific class among classes, and the number of the samples is counted for each class based on the feature information on the samples and the class label information on the samples, and determines whether the input pattern belongs to the specific class based on the feature information on the input pattern, the first conditional probability and the second conditional probability.