A non-parametric validation framework for photoplethysmography-based heart rate monitoring: a proof-of-concept study using the two-sample Kolmogorov–Smirnov test
[EN] This article introduces a statistical approach for comparing heart rate measurements obtained from two photoplethysmography (PPG) signals: one recorded with a commercial oximeter and the other acquired using a device based on photoplethysmography with synchronous detection. The study applies th...
| Autores: | , , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2026 |
| País: | España |
| Institución: | Universitat Politècnica de València (UPV) |
| Repositorio: | RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia |
| Idioma: | inglés |
| OAI Identifier: | oai:riunet.upv.es:10251/232181 |
| Acceso en línea: | https://riunet.upv.es/handle/10251/232181 |
| Access Level: | acceso abierto |
| Palabra clave: | Photoplethysmography Two-sample Kolmogorov Smirnov test Heart rate comparison Biomedical signal processing Non-invasive health monitoring |
| Sumario: | [EN] This article introduces a statistical approach for comparing heart rate measurements obtained from two photoplethysmography (PPG) signals: one recorded with a commercial oximeter and the other acquired using a device based on photoplethysmography with synchronous detection. The study applies the two-sample Kolmogorov Smirnov test as a robust and versatile method for comparing the distributions of PPG signals. By integrating the two-sample Kolmogorov Smirnov test into the validation process of cardiac pulse measurement devices, the work demonstrates its effectiveness in enhancing the accuracy and reliability of biomedical signal analysis. Results show that, when comparing signals against calibrated reference devices and visualizing cumulative distribution functions, the two-sample Kolmogorov Smirnov test is capable of detecting subtle differences in signal behavior. This innovative use of the two-sample Kolmogorov Smirnov test provides valuable insights for the design and validation of biomedical signal processing systems and contributes to the advancement of non-invasive health monitoring technologies. |
|---|