Background: With the rapid aging of the global population, early detection of dementia and Mild Cognitive Impairment (MCI) has become a critical public health issue. Current diagnostic tools often rely on costly or invasive methods that are difficult to implement in primary care settings. This study proposes a non-invasive screening model using Pulse Harmonic Analysis to evaluate physiological responses to cognitive load.

Methods: We recruited elderly participants categorized into an MCI group (n = 47) and a healthy control group (n = 11) based on Montreal Cognitive Assessment (MoCA) scores. Pulse wave signals were collected using a photoplethysmography (PPG) sensor before and after a cognitive intervention (subtraction task). Fast Fourier Transform (FFT) was used to extract harmonic amplitude (C-values), coefficient of variation (CV), phase (P-values), and phase standard deviation (PnSD) to analyze hemodynamic changes.

Results: The healthy control group exhibited a consistent downward trend in harmonic amplitudes (C1–C10) and stable phase characteristics following cognitive intervention, suggesting a resilient physiological regulation mechanism. In contrast, the MCI group demonstrated distinct signs of dysregulation: (1) High Heterogeneity: Amplitude responses were directionless and highly variable, with non-normal distributions; (2) Increased Instability: The Coefficient of Variation (CV) and Phase Standard Deviation (PnSD) generally increased after intervention, indicating a loss of autonomic stability; (3) Phase Sensitivity: Notably, the MCI group showed marginally significant changes in low-frequency phases (P1: p = 0.016 ; P2: p = 0.027 ), whereas the control group remained stable.

Conclusion: Although some amplitude parameters lacked statistical significance due to individual variability, the study identified critical physiological markers distinguishing MCI from healthy aging. Specifically, the significant alterations in low-frequency phases (P1, P2) and the increased signal instability (CV, PnSD) under cognitive load suggest that Pulse Harmonic Analysis can effectively capture cardiovascular-autonomic dysregulation. These features show promise as scalable, non-invasive biomarkers for early MCI screening.

Authors:
Kun-Yuan Hsiao, Ministry of Health and Welfare, Taiwan
Jia-Yu Wei, Ministry of Health and Welfare, Taiwan
Yen-Wen Chang, Ministry of Health and Welfare, Taiwan
Yong-Hong Chen, Ministry of Health and Welfare, Taiwan
Wan-Hsuan Huang, Ministry of Health and Welfare, Taiwan

About the Presenter(s)
Dr. Kun-Yuan Hsiao is currently an Attending Physician in the Department of Traditional Chinese Medicine at Cishan Hospital, Ministry of Health and Welfare, Taiwan. He possesses a comprehensive academic background, holding a Ph.D. from the Institute

Connect on Linkedin
https://www.linkedin.com/in/坤元-蕭-80ab7a392

See this presentation on the full schedule – Tuesday Schedule

Bookmark