QUMPHY Software package

At the core of this project stands the development of measures to quantify the uncertainties associated with machine learning (ML) algorithms applied to medical problems, in particular the analysis and processing of photoplethysmography (PPG) signals. To achieve this the following tasks will be addressed: (i) benchmark datasets will be generated using publicly available in vivo, in vitro and synthetic data (ii) different ML models and uncertainty quantification (UQ) methods will be used to analyse the processing of the PPG signals and specify the associated uncertainty and (iii) a good practice guide with accompanying software repository showcasing the used models, methods and benchmarks will be developed and made publicly available.

Accompanying Software

To ensure transparency and reproducibility, the QUMPHY project provides dedicated software repositories that accompany each of our major publications. These repositories, hosted on the project’s official GitLab group, contain the source code used to generate our results. By sharing these tools, we aim to provide the research community with the necessary assets to validate our findings and adapt our algorithms and benchmarks to their own applications.