Voluntary breathing (VB), short-term exercise (STE), and mental stress (MS) can modulate breathing rate (BR), heart rate (HR), and blood pressure (BP), thereby affecting human physical and mental state. While existing experimental studies have explored the relationship between VB, STE, or MS and BR, HR, and BP changes, their findings remain fragmented due to individual differences and challenges in simultaneous, BR, HR, and BP measurements. We propose a computational approach for in-silico simultaneous measurements of the physiological values by comprehensive prediction of the respiratory and circulatory system responses to VB, STE, or MS. Our integrated model combines a respiratory system with a circulatory model, leveraging actor-critic reinforcement learning to control respiratory muscles. We introduce specific parameters to account for involuntary or VB and hyperventilation induced by STE. We modeled mental stress as an electrical input to the amygdala based on prior studies indicating that stress leads to amygdala hyperactivity. Our predictions for breathing rate (BR), tidal volume, minute ventilation, HR, and BP are validated against literature data obtained during various conditions, including different VB patterns (ranging from 6 to 14 bpm), active or passive knee flexion STE, and MS load. The model demonstrates good agreement with experimental results and highlights its ability to explore the mechanism of individual differences. Our model predicts heart rate variability (HRV) indices of total power spectral density and the ellipse area of Poincaré plot. Notably, slow deep breathing at a BR of 6 bpm increases HRV indices, promoting relaxation and cognitive performance. Conversely, MS elevates BP but reduces HRV indices, indicating an unstable and risky state for mental and physical health. Overall, our proposed computational approach provides simultaneous and reasonable predictions of various physiological values, accounting for individual variations through specific parameters.
Read full abstract