ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2403.13658
18
1

Multimodal Variational Autoencoder for Low-cost Cardiac Hemodynamics Instability Detection

20 March 2024
M. N. I. Suvon
P. C. Tripathi
Wenrui Fan
Shuo Zhou
Xianyuan Liu
S. Alabed
V. Osmani
Andrew J Swift
Chen Chen
Haiping Lu
ArXivPDFHTML
Abstract

Recent advancements in non-invasive detection of cardiac hemodynamic instability (CHDI) primarily focus on applying machine learning techniques to a single data modality, e.g. cardiac magnetic resonance imaging (MRI). Despite their potential, these approaches often fall short especially when the size of labeled patient data is limited, a common challenge in the medical domain. Furthermore, only a few studies have explored multimodal methods to study CHDI, which mostly rely on costly modalities such as cardiac MRI and echocardiogram. In response to these limitations, we propose a novel multimodal variational autoencoder (CardioVAEX,G\text{CardioVAE}_\text{X,G}CardioVAEX,G​) to integrate low-cost chest X-ray (CXR) and electrocardiogram (ECG) modalities with pre-training on a large unlabeled dataset. Specifically, CardioVAEX,G\text{CardioVAE}_\text{X,G}CardioVAEX,G​ introduces a novel tri-stream pre-training strategy to learn both shared and modality-specific features, thus enabling fine-tuning with both unimodal and multimodal datasets. We pre-train CardioVAEX,G\text{CardioVAE}_\text{X,G}CardioVAEX,G​ on a large, unlabeled dataset of 50,98250,98250,982 subjects from a subset of MIMIC database and then fine-tune the pre-trained model on a labeled dataset of 795795795 subjects from the ASPIRE registry. Comprehensive evaluations against existing methods show that CardioVAEX,G\text{CardioVAE}_\text{X,G}CardioVAEX,G​ offers promising performance (AUROC =0.79=0.79=0.79 and Accuracy =0.77=0.77=0.77), representing a significant step forward in non-invasive prediction of CHDI. Our model also excels in producing fine interpretations of predictions directly associated with clinical features, thereby supporting clinical decision-making.

View on arXiv
Comments on this paper