University of Twente Student Theses
Physical Activity Energy Expenditure Estimation in Activities of Daily Living using a Simplified Physiological Computing Model
Poelarends, R.J. (2024) Physical Activity Energy Expenditure Estimation in Activities of Daily Living using a Simplified Physiological Computing Model.
PDF
6MB |
Abstract: | Introduction: With over one billion people suffering from obesity, addressing energy imbalance between calories consumed and expended is critical. A tailored and dynamic model is essential for providing personalized insights into energy expenditure patterns, especially for development of personalized obesity interventions. In this study, we aimed to develop a simplified physiological computing model to accurately estimate physical activity energy expenditure (PAEE). Methods: An observational study was conducted at the eHealth House, University of Twente, involving 10 participants aged 23 - 49. Participants performed various activities of daily living monitored by multiple wearable sensors. A simplified three-compartment model was utilized, incorporating the lungs, circulation system, and muscle tissue. The model estimated PAEE using heart rate and inertial measurement unit (IMU) data as inputs. Two model variations were tested: model 1 utilized pelvis IMU data, and model 2 integrated IMU data from both the pelvis and thighs. Performance metrics were coefficient of determination (R2) and root mean square error (RMSE). Results: Model 1 exhibited a significantly lower mean R2 compared to model 2 (0.256 ± 0.302 vs. 0.375 ± 0.312, p=0.05). No significant difference was found between model 1 and model 2 in terms of RMSE (0.112 ± 0.035 kJ/min/kg vs. 0.100 ± 0.037 kJ/min/kg, p = 0.08). In terms of RMSE, both models performed significantly worse on the cycling activity for each participant, compared to the other activities. Discussion/Conclusion: In this study, we presented a promising approach for PAEE estimation using a simplified physiological computing model, accounting for individual physiological differences and the dynamics of energy expenditure. By integrating wearable sensor data with physiological principles, our method might offer a significant advancement in personalized health monitoring and obesity research, paving the way for more effective interventions and lifestyle improvements. |
Item Type: | Essay (Master) |
Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
Subject: | 44 medicine |
Programme: | Biomedical Engineering MSc (66226) |
Link to this item: | https://purl.utwente.nl/essays/101567 |
Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Repository Staff Only: item control page