Electroencephalographic Insights into Applied Pressure Physiotherapy: A Wavelet-Based Machine Learning Approach

Abstract

The current research uses EEG signal analysis to investigate how Applied Pressure Physiotherapy (APP) modulates brain activity. Before and after APP sessions, 15 volunteers had their EEGs recorded. Prior to Maximal-Overlap Discrete Wavelet Packet Transform (MODWPT) decomposition spanning the Delta, Theta, Alpha, Beta, and Gamma bands, pre-processing comprised band-pass filtering (0.5–40 Hz) and artifact removal. Significant variations in power were observed after APP: Alpha power climbed by 4.8%, which indicates a more alert but relaxed condition, whilst Delta and Theta bands declined with 9.4% and 7.8%, respectively. Decreased cognitive stress is further demonstrated by beta and gamma power drops for 9.8% and 6.7%, respectively. Neural stability improvements post APP have been observed by Hjorth measures, such as Activity (−11.5%), Mobility (−45.3%), as well as Complexity (−41.5%). Utilizing parameters like Mean Curve Length (MCL), classification analyses with Linear Discriminant Analysis (LDA) and Random Forest (RF) acquired excellent precision (99.97%). These results emphasize APP's viability as a non-invasive treatment promoting mental wellness and cognitive improvement by demonstrating that it induces a relaxed mental state.