Subject-Specific Cardiovascular Signatures Revealed by Arterial Pulse Signals in Patients with Heart Diseases—Part I: General Cases

Abstract

Aim: This study evaluates whether cardiovascular (CV) parameters derived from noninvasive arterial pulse signals can capture subject-specific physiological characteristics and longitudinal changes in patients with diverse heart diseases. Method: Arterial pulse signals at the radial artery (RA) and carotid artery (CA) were recorded using microfabricated tactile sensors and analyzed with a single-degree-of-freedom time–frequency (SDOF-TF) framework. Key CV parameters, including heart rate (HR), heart rate variability (HRV), respiration rate (RR), respiration modulation (RM), harmonic amplitudes, and normalized arterial pulse waveform (APW), were extracted from measurements at-rest and post-exercise across three visits in eighteen patients with heterogeneous CV conditions. Longitudinal changes in CV parameters within subjects and variability between subjects were assessed. Results: Compared to a healthy reference, patients exhibited altered CV profiles indicative of impaired autonomic regulation. Common trends included reduced HRV, blunted post-exercise CV responses, and attenuation of higher-order harmonics at the RA. CA-derived APW features showed enhanced sensitivity to systolic upstroke slope compared to the RA. Marked inter-subject variability was evident, reflected in at-rest values of key CV parameters and 5min post-exercise responses, highlighting the need for personalized interpretation. Longitudinal measurements revealed mixed effects of exercise on the CV system, likely influenced by concurrent medication and the short observation period. Conclusion: The SDOF-TF framework applied to arterial pulse signals effectively reveals both shared and individualized CV dysfunctions, capturing qualitative harmonic signatures, site-specific differences, and complex parameter interactions not detected by conventional methods, offering potential improvements in diagnosis and treatment monitoring for patients with diverse CV diseases.

The authors' abstract, as published at the source. Diagnostics, 2026 · DOI ↗