Unobtrusive ultra-short-term HRV analysis for the identification of physiological states in healthy people
(English) In recent years, heart rate variability (HRV) has become an essential tool for assessing the function of the autonomic nervous system (ANS), reflecting the balance between the sympathetic (SNS) and parasympathetic (PNS) systems. A high HRV indicates a healthy ANS and better adaptability to...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis doctoral |
| Estado: | Versión publicada |
| Fecha de publicación: | 2025 |
| País: | España |
| Institución: | CBUC, CESCA |
| Repositorio: | TDR. Tesis Doctorales en Red |
| OAI Identifier: | oai:www.tdx.cat:10803/694898 |
| Acceso en línea: | http://hdl.handle.net/10803/694898 https://dx.doi.org/10.5821/dissertation-2117-433757 |
| Access Level: | acceso abierto |
| Palabra clave: | HRV (Heart Rate Variability) Ultra-short-term HRV Physiological states analysis Unobtrusive health monitoring Àrees temàtiques de la UPC::Enginyeria biomèdica Àrees temàtiques de la UPC::Informàtica 004 - Informàtica 616.1 - Patologia del sistema circulatori dels vasos sanguinis. Trastorns cardiovasculars 612 - Fisiologia |
| Sumario: | (English) In recent years, heart rate variability (HRV) has become an essential tool for assessing the function of the autonomic nervous system (ANS), reflecting the balance between the sympathetic (SNS) and parasympathetic (PNS) systems. A high HRV indicates a healthy ANS and better adaptability to stress. It is measured non-invasively using electrocardiography (ECG) or photoplethysmography (PPG), providing information about cardiovascular health, stress responses, and physiological resilience. Factors such as exercise, metabolic processes, and recovery influence HRV, making it useful for characterizing human regulatory systems. HRV is also used in the sports field to optimize performance through personalized training. Analysis methods include time-domain, frequency-domain, and non-linear techniques. However, the growing demand for real-time assessments has driven the development of very short-term HRV analysis (less than 5 minutes), facilitated by wearable devices such as smartwatches and fitness trackers. These devices allow real-time health monitoring through on-demand self-assessments, improving quality of life. Wearable technology, such as smartwatches and health monitors, has grown significantly, offering real-time tracking of biosignals, physical activity, and sleep patterns. Nonetheless, challenges persist due to a lack of transparency in the methods used to calculate HRV in these devices. This thesis addresses these challenges by developing techniques for very short-term HRV assessments, focusing on monitoring health status and physiological changes in healthy individuals, including tracking sleep stages, arousal states, and postural changes. Accurate heartbeat detection is essential for reliable HRV calculations, which are critical for both clinical diagnostics and consumer health applications. This requires identifying QRS peaks in ECG signals or pulses in PPG signals and improving quality through preprocessing and feature extraction. This dissertation explores algorithms to optimize the accuracy of heartbeat detection, advancing HRV evaluation and cardiac monitoring in wearable devices. HRV analysis begins with heartbeat detection, using algorithms based on signal morphology and inter-beat interval features. Designing reliable beat detectors for ambulatory environments presents a challenge, as they must be non-invasive and effective in real-world scenarios. Therefore, one of the main objectives is to improve the accuracy and reliability of beat detection, thereby optimizing HRV evaluation and cardiac monitoring. This thesis addresses two main challenges: (1) reliable heartbeat detection in ambulatory environments using wearable sensors and (2) very short-term HRV measurements to identify physiological states in healthy individuals. Accurate detectors and analytical techniques are developed to improve the precision of beat detection and the identification of physiological states. In conclusion, the thesis proposes a theoretical framework for non-invasive methods of HRV estimation and monitoring physiological changes through ultra-short-term analysis. It details materials, methods, and novel techniques, emphasizing the relevance of these approaches to understanding autonomic nervous system function and cardiovascular health. This thesis aims to make a significant contribution to the field of mobile health (m-health), enabling reliable monitoring in everyday life. |
|---|