Physics-Based PPG Simulation for Chronic Kidney Fluid Retention Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting Chronic Kidney Disease (CKD) lack efficiency and accuracy, and existing digital biomarkers face challenges with privacy and cost issues, making early intervention difficult.
Innovation Solution
A processor-implemented method using Photoplethysmography (PPG) signals, physics-based modeling, and a PlethAugment-based conditional generative adversarial network (CGAN) model to detect CKD by simulating PPG signals and training a discriminator for accurate CKD detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional techniques are used for detecting CKD, then the detection process is simple, but the efficiency and accuracy are insufficient
Solution Approach 1:
The patent creates a virtual copy of PPG signals through physics-based simulation, generating synthetic training data that replicates real physiological signals. This allows the development of accurate detection algorithms without requiring extensive real patient data, thereby improving detection accuracy while managing data acquisition complexity
Solution Approach 2:
The patent performs preliminary data preparation by generating synthetic PPG signals and training the CGAN model in advance. This preliminary action creates a pre-trained detection system that can be deployed without requiring real-time complex data processing during actual CKD detection, resolving the contradiction between accuracy and operational complexity
2Measurement precision
If patient health data is collected for CKD research, then detection accuracy can be improved, but privacy and security issues arise
Solution Approach 1:
The patent replaces real patient health data with synthetically generated PPG signals that preserve the statistical and physiological characteristics of real data. This copying approach maintains detection accuracy while completely eliminating privacy and security risks associated with real patient information
Solution Approach 2:
The physics-based simulation model acts as an intermediary between the need for accurate training data and the requirement for patient privacy. It generates intermediate synthetic data that serves the training purpose without exposing actual patient information, thus resolving the contradiction between accuracy improvement and privacy protection
3Reliability
If digital biomarkers are developed for CKD detection, then early intervention becomes possible, but cost increases for organizations and patients
Solution Approach 1:
The patent uses synthetic signal generation to replace expensive real-world data collection and annotation processes. By copying the essential characteristics of PPG signals through physics-based models, it reduces the cost of developing and implementing digital biomarkers while maintaining early detection reliability
Solution Approach 2:
The patent employs computationally efficient CGAN models that can be trained once with synthetic data and then deployed repeatedly without additional cost. This approach replaces continuous expensive data acquisition with a one-time training investment, making the digital biomarker solution more cost-effective for organizations and patients
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method provides efficient and accurate detection of CKD by generating authentic PPG signals, enabling effective fluid retention analysis in Chronic Kidney Care.
Implementation Method 1
receiving (i) a data associated with one or more Photoplethysmography (PPG) based biomarkers
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
The disclosure relates generally to methods and systems for fluid retention analysis in Chronic Kidney Care. Conventional techniques for detecting the CKD in the subject lack with efficiency and accuracy since most of the ailments are silent and variant in nature pinpointing a universal cause or biomarker is difficult. The methods and systems of the present disclosure propose an in-silico model-based approach to detect CKD in the subject. In the first stage, a standard PPG waveform is simulated using the physicsbased model. In the second stage, a Generative Adversarial Network (GAN) model is trained using the simulated PPG data and the reference experimental PPG data. In the third and the last stage, the discriminator model of the trained GAN model is employed to evaluate and analyze the test dataset of the subject whose CKD is to be detected.