INN-PAR: Invertible Neural Network for PPG to ABP Reconstruction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kundu, Soumitra, Panda, Gargi, Bhattacharya, Saumik, Routray, Aurobinda, Guha, Rajlakshmi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912275759104000
author Kundu, Soumitra
Panda, Gargi
Bhattacharya, Saumik
Routray, Aurobinda
Guha, Rajlakshmi
author_facet Kundu, Soumitra
Panda, Gargi
Bhattacharya, Saumik
Routray, Aurobinda
Guha, Rajlakshmi
contents Non-invasive and continuous blood pressure (BP) monitoring is essential for the early prevention of many cardiovascular diseases. Estimating arterial blood pressure (ABP) from photoplethysmography (PPG) has emerged as a promising solution. However, existing deep learning approaches for PPG-to-ABP reconstruction (PAR) encounter certain information loss, impacting the precision of the reconstructed signal. To overcome this limitation, we introduce an invertible neural network for PPG to ABP reconstruction (INN-PAR), which employs a series of invertible blocks to jointly learn the mapping between PPG and its gradient with the ABP signal and its gradient. INN-PAR efficiently captures both forward and inverse mappings simultaneously, thereby preventing information loss. By integrating signal gradients into the learning process, INN-PAR enhances the network's ability to capture essential high-frequency details, leading to more accurate signal reconstruction. Moreover, we propose a multi-scale convolution module (MSCM) within the invertible block, enabling the model to learn features across multiple scales effectively. We have experimented on two benchmark datasets, which show that INN-PAR significantly outperforms the state-of-the-art methods in both waveform reconstruction and BP measurement accuracy. Codes can be found at: https://github.com/soumitra1992/INNPAR-PPG2ABP.
format Preprint
id arxiv_https___arxiv_org_abs_2409_09021
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle INN-PAR: Invertible Neural Network for PPG to ABP Reconstruction
Kundu, Soumitra
Panda, Gargi
Bhattacharya, Saumik
Routray, Aurobinda
Guha, Rajlakshmi
Machine Learning
Human-Computer Interaction
Non-invasive and continuous blood pressure (BP) monitoring is essential for the early prevention of many cardiovascular diseases. Estimating arterial blood pressure (ABP) from photoplethysmography (PPG) has emerged as a promising solution. However, existing deep learning approaches for PPG-to-ABP reconstruction (PAR) encounter certain information loss, impacting the precision of the reconstructed signal. To overcome this limitation, we introduce an invertible neural network for PPG to ABP reconstruction (INN-PAR), which employs a series of invertible blocks to jointly learn the mapping between PPG and its gradient with the ABP signal and its gradient. INN-PAR efficiently captures both forward and inverse mappings simultaneously, thereby preventing information loss. By integrating signal gradients into the learning process, INN-PAR enhances the network's ability to capture essential high-frequency details, leading to more accurate signal reconstruction. Moreover, we propose a multi-scale convolution module (MSCM) within the invertible block, enabling the model to learn features across multiple scales effectively. We have experimented on two benchmark datasets, which show that INN-PAR significantly outperforms the state-of-the-art methods in both waveform reconstruction and BP measurement accuracy. Codes can be found at: https://github.com/soumitra1992/INNPAR-PPG2ABP.
title INN-PAR: Invertible Neural Network for PPG to ABP Reconstruction
topic Machine Learning
Human-Computer Interaction
url https://arxiv.org/abs/2409.09021