SkillAgentSearch skills...

T2TGAN

This repository is for our work submitted to ArXiv and under review at EMBC2021 titled "Estimation of Continuous Blood Pressure from PPG via a Federated Learning Approach".

Install / Use

/learn @Brophy-E/T2TGAN
About this skill

Quality Score

0/100

Supported Platforms

Universal

README

T2TGAN

This repository is for our work submitted to ArXiv and journal paper under review titled "Estimation of Continuous Blood Pressure from PPG via a Federated Learning Approach".

We implement a framework that is capable of inferring ABP from a single optical photoplethysmogram (PPG). We train our global-model across distributed client-models and data sources to mimic a large-scale distributed collaborative learning experiment that could be implemented across low-cost wearables. Our T2TGAN can map any given time series signal to another but for the transform to make sense the signals should be correlated in some way.

GAN Architecture used in this work: P2B_Architecture

The TimeSeries-to-TimeSeries Architecture: CycleGAN

Federated Learning setup in this repo: FedGAN

Below is an example of the generated data: RealandFakeABP

Related Skills

View on GitHub
GitHub Stars25
CategoryEducation
Updated1y ago
Forks7

Languages

Jupyter Notebook

Security Score

60/100

Audited on Feb 26, 2025

No findings