Generative Network Models for Realistic Physiological Feature Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for evaluating pharmacokinetic behaviors of drugs face challenges due to insufficient physiological feature data samples and discrepancies between simulated and real physiological feature data distributions, leading to inaccurate processing.
Innovation Solution
A data processing method utilizing a target network model for data generalization or prediction to generate or predict physiological feature data, incorporating models like generative adversarial networks, variational autoencoders, or diffusion models to enhance data availability while conforming to real data distributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed equations based on prior knowledge are used to establish virtual individual groups, then the evaluation process is simplified, but the generated physiological feature data distribution deviates significantly from real-life distributions
Solution Approach 1:
The patent uses generative adversarial networks to copy the distribution characteristics of real physiological feature data. The generator network learns to replicate the complex patterns and correlations in real data, producing synthetic data that maintains authentic distribution properties while avoiding the oversimplification of fixed equations.
Solution Approach 2:
The patent transforms the approach by changing from fixed deterministic equations to probabilistic generative models. This parameter change allows the system to capture the variability and uncertainty inherent in physiological data, generating diverse yet realistic samples that reflect true population distributions.
2Quantity of substance
If more physiological feature data samples are collected to improve evaluation accuracy, then data sufficiency increases, but data acquisition time and cost increase
Solution Approach 1:
The patent performs preliminary action by training the generative model on available real data beforehand. Once trained, the model can rapidly generate additional synthetic samples without requiring further time-consuming data collection, thus preparing the system in advance to provide sufficient data for evaluation when needed.
Solution Approach 2:
Instead of collecting more real data through time-consuming measurements, the system creates copies of existing data through the generative model. These synthetic copies preserve the statistical properties and correlations of real physiological data, providing sufficient sample quantities for robust pharmacokinetic evaluation.
Data Source
AI summary
Provided are a data processing method, an electronic device, and a storage medium. The method includes the following: The physiological feature data of a target object under at least one physiological indicator is acquired; a data processing type corresponding to the physiological feature data is determined, and a target network model corresponding to the data processing type is invoked; and the physiological feature data is processed based on the target network model to obtain target physiological feature data.