Benchmark Problems and Benchmark Datasets for the evaluation of Machine and Deep Learning methods on Photoplethysmography signals: the D4 report from the QUMPHY project

2026-04-01

AI Summary

This work addresses the current lack of standardized benchmarks for evaluating machine learning and deep learning methods in photoplethysmography (PPG) signal analysis. To bridge this gap, the study systematically establishes the first standardized benchmark framework tailored to PPG analysis, defining six clinically relevant medical tasks as core evaluation problems. Accompanying this framework, the authors release a high-quality benchmark dataset along with comprehensive usage guidelines. Crucially, the proposed benchmark incorporates uncertainty quantification mechanisms to enable fair and reliable performance assessment of algorithms. This contribution provides a robust, reproducible foundation that significantly advances standardized research and development of medical AI models for PPG-based applications.

Abstract

This report is part of the Qumphy project (22HLT01 Qumphy) that is funded by the European Union and is dedicated to the development of measures to quantify the uncertainties associated with Machine Learning algorithms applied to medical problems, in particular the analysis and processing of Photoplethysmography (PPG) signals. In this report, a list of six medical problems that are related to PPG signals and serve as Benchmark Problems is given. Suitable Benchmark datasets and their usage are described also.

Problem

• Photoplethysmography
• Benchmark Problems
• Benchmark Datasets
• Machine Learning
• Uncertainty Quantification

Innovation

• Photoplethysmography
• Benchmark Problems
• Benchmark Datasets
• Uncertainty Quantification
• Machine Learning