Independent component analysis (ICA) and neural networks were used to extract sub-signals of heart rate variability (HRV). Electrocardiogram (ECG) recording was consisted of six minutes when the volunteer was lying and another six minutes when the volunteer was standing. HRV was extracted from ECG. According to time-delay, HRV was divided into five groups as mixed signals. ICA and neural networks reconstructed five signals into two groups. Results showed that one group signal component centralized in low frequency (called IC1); the other did in high frequency (called IC2). The power of IC1 was significantly increased (P<0.05) and the ratio of IC1 to total power was significantly increased (P<0.01) from lying to standing. Comparing the two postural results, it shows that IC1 may express sympathetic activity, and IC2 represents parasympathetic activity. Sympathetic and parasympathetic nervous function can be evaluated respectively and quantificationally by data and graphs from the two decomposed components. Key word: Neural networks, Independent component analysis, Heart rate variability, Autonomic nervous system.

No abstract