High-precision personalized respiratory guidance model for enhanced breathing training: effects on heart rate variability

Abstract

Objectives

While slow, controlled breathing at 6 breaths per minute(bpm) has been shown to enhance heart rate variability (HRV), the impact of key parameters such as the inhalation-to-exhalation ratio (IER) remains unclear. This study aims to develop a scientifically valid and broadly applicable respiratory model to improve the accuracy of controlled breathing training systems, explore the relationship between breathing parameters and HRV, and assess their effects on autonomic nervous system regulation.

Methods

A high-precision, personalized respiratory training system was developed, incorporating precise visual and auditory guidance based on a feature-fitting model using B-spline fitting and particle swarm optimization. The system adaptively generated breathing guidance according to individual respiratory features, including rate, depth, and IER. Ten healthy participants each completed 20 sessions with different distinct patterns while HRV indicators were monitored.

Results

The model exhibited a mean fitting error below 0.1, with 96 % of cycles closely matching target patterns, demonstrating effective and reliable training guidance. Breathing at 6 bpm with an IER of 0.5 yielded the highest HRV. Additionally, 4–7–8 and box breathing patterns also significantly enhanced HRV.

Conclusion

This study proposed a scientifically valid, high-precision respiratory guidance model for personalized training. It also demonstrates that slow breathing at 6 bpm significantly enhances vagal activity and parasympathetic tone. Furthermore, a lower IER was associated with increased HRV, implying that optimizing this ratio can further improve outcomes.