Work

Direct Estimation of Pediatric Heart Rate Variability from BOLD-fMRI: A Machine Learning Approach Using Dynamic Connectivity

Year: 2025

Type: conference-paper

Abstract: Motivation: In many pediatric fMRI studies, cardiac signals are missing or have poor quality. Therefore, it would be highly beneficial to have a tool to extract Heart Rate Variation (HRV) waveforms directly from fMRI data without the need for peripheral recording devices. Goal(s): Develop a machine learning framework to accurately reconstruct HRV tailored for pediatric applications. Approach: A hybrid model using one-dimensional Convolutional Neural Networks (1D-CNN) and Gated Recurrent Units (GRU) analyzed BOLD signals from 628 ROIs, integrating past and future data. Results: Achieved an 8% improvement in HRV accuracy, evidenced by enhanced performance metrics, demonstrating the model's effectiveness. Impact: This method enhances pediatric fMRI by eliminating the need for peripheral photoplethysmography devices, reducing costs and simplifying procedures. It could also improve the robustness of pediatric fMRI studies, which are more affected by physiological and developmental variations than in adults.

Source: Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition

DOI: 10.58530/2025/3554

Authors: Abdoljalil Addeh, Karen Ardila, Rebecca J. Williams, G. Bruce Pike, M. Ethan MacDonald

Language: English

Topic: Heart Rate Variability and Autonomic Control

Subfield: Cardiology and Cardiovascular Medicine

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: green

OpenAlex ID: W4414236206

FWCI:

Cites:

Cited by:

Related to: