University of Twente Student Theses

Login

Stress Detection through Machine Learning using HRV : A Systematic Review

Fraters, Steven (2025) Stress Detection through Machine Learning using HRV : A Systematic Review.

[img] PDF
397kB
Abstract:Stress has a significant impact on the health of humans, thus reliable detection methods are required so we can prevent the ever increasing stressors in life from harming us. One of the current stress detection methods is looking at the biomarker Heart Rate Variability (HRV) measured by wearables. HRV is the difference between time intervals of separate heart beats, which decreases when stress is introduced to the body hence making it viable as a biomarker in stress prediction. Therefore this systematic review aims to investigate machine learning models that have been used in stress prediction with through HRV. The databases selected for this systematic review are Scopus, PubMed and IEE Xplore. Applying the criteria of articles that use HRV and machine learning in stress prediction resulted in 19 articles being included in the analysis. Analysis shows that the most commonly used machine learning algorithms are random forest, k-nearest neighbor and support vector machines. The challenges most models faced are with the personal differences in HRV changes people experience, as this makes it hard to make a general model. The most accurate models were Convolutional Neural Networks (CNN) and thus are my recommendation for an algorithm in stress prediction through HRV. Further research into CNN seems promising for people to monitor, and possibly regulate, their stress levels
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:https://purl.utwente.nl/essays/105250
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page