University of Twente Student Theses


Measuring pleasantness of bicyclists through heart rate variability over self-reports

Tobé, Mathijs (2023) Measuring pleasantness of bicyclists through heart rate variability over self-reports.

[img] PDF
Abstract:Ultra-short-term (UST) heart-rate-variability (HRV) analysis in non-static conditions, such as bicycling, poses significant challenges. Most HRV analysis is traditionally conducted in a resting state, calling for an examination of UST HRV metrics applicable to our specific setting. We are trying to estimate the axis of valence in the circumplex model of affect, which relates to pleasantness. Consequently, certain HRV metrics may exhibit unreliability or insignificance in this estimation. Considering software and hardware limitations, we focus on time-domain HRV metrics that can be derived from R-R intervals. To facilitate comparison with other subjects' data, R-R intervals are normalized by dividing each interval by the average R-R interval of that participant. In our experimental study, involving a sample size of N=23, we collected R-R intervals using a Polar H10 heart rate sensor and obtained self-reported pleasantness ratings from cyclists on a scale ranging from 1 (unpleasant) to 3 (pleasant), utilizing an experimental e-bike setup. Our aim is to assess the suitability of heart rate variability analysis for measuring pleasantness. The collected data undergoes preprocessing using two approaches, including a sliding window method. We analyze the data utilizing Support Vector Machines (SVMs) with various kernels and Random Forest classification. The highest achieved F1-score is 47% when using a window size of 130 R-R intervals and incorporating time-domain HRV features such as SDNN, RMSSD, NN50, and pNN50. These findings suggest that additional markers, including frequency-domain HRV and qualitative data, are crucial for a comprehensive assessment. Furthermore, the inclusion of wearable sensor data, such as electrodermal activity, respiration, and skin temperature, may be necessary to accurately estimate self-reported pleasantness in bicyclists.
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:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page