SCG With Your Phone: Diagnosis of Rhythmic Spectrum Disorders in Field Conditions

TL;DR

This paper introduces a deep learning-based method for accurate detection of aortic valve opening events in noisy smartphone-recorded seismocardiography signals, enabling practical mobile-based cardiac rhythm analysis in real-world conditions.

Contribution

It presents an enhanced U-Net v3 architecture with novel adaptive projection and post-processing techniques for robust SCG segmentation across diverse devices and noisy environments.

Findings

High accuracy in AO detection across multiple smartphone models

Robust performance in noisy, real-world data collection scenarios

Enables low-cost, automated cardiac monitoring with smartphones

Abstract

Aortic valve opening (AO) events are crucial for detecting frequency and rhythm disorders, especially in real-world settings where seismocardiography (SCG) signals collected via consumer smartphones are subject to noise, motion artifacts, and variability caused by device heterogeneity. In this work, we present a robust deep-learning framework for SCG segmentation and rhythm analysis using accelerometer recordings obtained with consumer smartphones. We develop an enhanced U-Net v3 architecture that integrates multi-scale convolutions, residual connections, and attention gates, enabling reliable segmentation of noisy SCG signals. A dedicated post-processing pipeline converts probability masks into precise AO timestamps, whereas a novel adaptive 3D-to-1D projection method ensures robustness to arbitrary smartphone orientation. Experimental results demonstrate that the proposed method achieves consistently high accuracy and robustness across various device types and unsupervised data-collection conditions. Our approach enables practical, low-cost, and automated cardiac-rhythm monitoring using everyday mobile devices, paving the way for scalable, field-deployable cardiovascular assessment and future multimodal diagnostic systems.