JMIR Publications

ABSTRACT

Background:

Continuous vital sign monitoring ensures early detection, prevents ICU admissions, and improves patient outcomes. Continuous heart rate (HR) monitoring methods often require direct skin contact, which can lead to patient discomfort. The rising popularity of ballistocardiography (BCG) offers a promising, non-contact solution for continuous vital sign monitoring with improved patient comfort.

Objective:

This study aims to develop and validate a novel HR measurement algorithm leveraging convolutional neural networks (CNNs) and ballistocardiography (BCG) signals. for accurate, non-contact, continuous HR monitoring. By integrating time-domain peak detection with Short-Time Fourier Transform (STFT) and CNN models, the proposed approach seeks to enhance HR measurement accuracy across diverse healthcare settings. The study follows the US-FDA’s Good Machine Learning Practice guidelines and evaluates the algorithm’s robustness, generalizability, and clinical applicability through extensive testing on a diverse dataset, ensuring improved patient comfort and early detection of clinical deterioration.

Methods:

The proposed algorithm combines time-domain peak detection with Short-Time Fourier Transform (STFT) and Convolutional Neural Networks (CNN) to enhance HR measurement from BCG signals. The CNN model developed was trained on 129,976 data points from 373 subjects (HR range: 36–230 bpm), including ICU patients, and tuned on 75,970 data points from 192 subjects (HR range: 46–169 bpm), with HR obtained from clinical-grade ECG devices, to improve generalizability. The algorithm was tested on 70,211 data points from 205 subjects, including ICU patients, across five independent studies, to demonstrate robust performance against diverse settings, demographics and comorbidities. The methodology is in compliance with US-FDA’s Good Machine Learning Practice for Medical Device Development: Guiding Principles

Results:

The algorithm achieved a Mean Absolute Error (MAE) of under 3 bpm and a Detection Rate (DR) exceeding 80%, underscoring its robustness. The Bland-Altman analysis indicates high accuracy with a minimal bias of 0.25 and limits of agreement within 8.59 bpm. Additionally, a Pearson’s correlation coefficient of 0.97 from the Deming regression further demonstrates strong alignment with reference HR measurements, reinforcing its precision and reliability for clinical applications

Conclusions:

This CNN-based algorithm presents a robust solution for contactless HR monitoring, addressing limitations of prior methods in noise management and adaptability. Its demonstrated accuracy, particularly in real-world, noisy clinical environments, highlights its potential for broad application in patient monitoring, improving comfort.