Monitoring cardiac function during physical activities is crucial for preventing serious events like heart attacks or arrhythmias. Evaluating cardiac responses at rest and after exercise provides important insights into cardiovascular health and fitness. At rest, metrics such as resting heart rate, blood pressure, and heart rhythm give an understanding of baseline cardiac function and potential underlying conditions. Post-exercise, the analysis of heart rate recovery, blood pressure changes, and heart rate variability highlights the cardiovascular system’s efficiency in responding to physical stress. This study aims to classify two physiological states, rest and post-exercise, based on cardiac responses measured with Finapres technology. Key cardiac metrics, including heart rate, systole, diastole, stroke volume, and pulse interval signals, were recorded over a 5-minute period. A 1D convolutional neural network (CNN) was used for classification, leveraging multiple features of each parameter. The results confirm the proposed method’s effectiveness in accurately distinguishing between rest and post-exercise states based on cardiac response features extracted via blood pressure continous monitor in a unobtrusive way. This research advances non-invasive techniques for monitoring cardiovascular health and evaluating physiological responses to exercise.

--

Finometer,Heart rate,Blood pressure,Convolutional neural network (CNN),Deep learning

• Computer and Information Sciences - Natural Sciences
• Electrical Engineering, Electronic Engineering, Information Engineering - Engineering and Technology
• Medical Engineering - Engineering and Technology