Key result
Evolutionary learning zero-poles model accurately fits exercise heart rate and identifies faster than least-squares.
Why the study?
Emerging digital phenotyping approaches require individualized model estimation of heart rate response to physical activity.
Does a zero-poles model with data-driven evolutionary learning improve the modeling and identification of heart rate dynamics compared to existing models?
Observational (n=30)
Does a zero-poles model with data-driven evolutionary learning improve the modeling and identification of heart rate dynamics compared to existing models?
Effect estimate: Pearson's ρ >0.95
p-value: p=<.03
A novel zero-poles model with evolutionary learning accurately and rapidly models heart rate dynamics during exercise, offering potential parameters for digital biomarkers.
May facilitate rapid heart rate modeling; leaves open validation as digital biomarkers.
Heart rate response to physical activity is widely investigated in clinical and training practice, as it provides information on a person's physical state. For emerging digital phenotyping approaches, there is a need for individualized model estimation. In this study, we propose a zero-poles model and a data-driven evolutionary learning method for identification. We also perform a comparison with existing first and second order models and gradient descent identification methods. The proposed model is based on a five-phase description of heart rate dynamics. Data was collected from 30 healthy participants using a treadmill and a thoracic sensor in two protocols (static and dynamic), for increasing and decreasing activity. Results show that the zero-poles model is a good fit for heart rate response to exercise (Pearson's coefficient ρ>.95), while first and second order models are also suitable (ρ>.92). The evolutionary learning method shows excellent results for fast model identification, in comparison with least-squares methods (p<.03). We surmise that the parameters of investigated linear dynamic models make good candidates for digital biomarkers and continuous monitoring.
No takes yet. Share an insight, caveat, or question.
Pătrașcu et al. (2024) conducted an observational in Healthy (n=30). Zero-poles model with data-driven evolutionary learning vs. First and second order models and least-squares methods was evaluated on Model fit for heart rate response to exercise and speed of model identification (Pearson's ρ >0.95, p=<.03). A zero-poles model with evolutionary learning provided a good fit for heart rate response to exercise (ρ>0.95) and faster identification than least-squares methods (p<0.03).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: