Joint Clinical Data Classification And Regression Using Shared Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for analyzing clinical data using machine learning separately for classification and regression result in low accuracy due to the lack of consideration for the correlation between disease status related parameters and classification results, leading to poor prediction performance.
Innovation Solution
A method and system that utilizes joint training of regression and classification models, where a neural network extracts feature information, a regression model predicts a disease status parameter, and a classification model predicts a disease status classification result, with the models' outputs being fused to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate classification and regression models are used independently, then the system complexity is low and ease of operation is maintained, but the prediction accuracy and performance are poor
Solution Approach 1:
The patent merges separate classification and regression models into a unified joint training framework where both models share common feature extraction components and are trained simultaneously on the same dataset. This allows the models to leverage correlated information from both task types, improving prediction accuracy while maintaining manageable complexity through shared architecture.
Solution Approach 2:
The patent creates a multi-functional system where a single joint training framework performs both classification and regression tasks simultaneously. The shared feature extraction and modeling components serve multiple purposes, extracting features that are useful for both disease status classification and continuous parameter prediction, thereby improving overall prediction accuracy.
2Reliability
If separate classification and regression models are used independently, then the training process is simple and fast, but the correlation between disease status parameters and classification results is not considered
Solution Approach 1:
The patent implements continuous joint training where classification and regression models are trained simultaneously in an integrated framework rather than sequentially. This continuous interaction during training allows the models to continuously leverage correlated information from both task types, improving prediction reliability while the efficient shared architecture minimizes additional training time.
Data Source
AI summary
This disclosure discloses a method for analyzing clinical data. The Method includes extracting a first feature information by applying a neural network to the clinical data; predicting a disease status related parameter by applying a regression model to the extracted first feature information; generating a second feature information based on the extracted first feature information and the disease status related parameter; and predicting a disease status classification result by applying a classification model to the second feature information. The method can improve the prediction accuracy and the diagnosis efficiency of doctors.