Summary of Interpretable Multimodal Learning For Cardiovascular Hemodynamics Assessment, by Prasun C Tripathi et al.
Interpretable Multimodal Learning for Cardiovascular Hemodynamics Assessment
by Prasun C Tripathi, Sina Tabakhi, Mohammod N I Suvon, Lawrence Schöb, Samer Alabed, Andrew J Swift, Shuo Zhou, Haiping Lu
First submitted to arxiv on: 6 Apr 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed multimodal learning pipeline predicts Pulmonary Arterial Wedge Pressure (PAWP) marker, a crucial cardiovascular hemodynamics marker for detecting heart failure. The pipeline combines Cardiac Magnetic Resonance Imaging (CMR) scans and Electronic Health Records (EHRs), leveraging tensor-based learning to extract spatio-temporal features from CMR scans and graph attention networks to select important EHR features. Four feature fusion strategies are designed: early, intermediate, late, and hybrid fusion. A linear classifier with interpretable results is employed. The pipeline is validated on a large dataset of 2,641 subjects from the ASPIRE registry, outperforming state-of-the-art methods through comparative studies. Decision curve analysis further confirms the pipeline’s applicability for screening a large population. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers developed a new way to predict PAWP marker, which is important for diagnosing heart failure. They used special scans and patient data to create a model that can predict this important measurement. The model combines information from different sources, such as heart scans and patient records, and uses advanced techniques like graph networks and feature fusion. The researchers tested their model on a large dataset of 2,641 patients and found it outperformed other methods. |
Keywords
» Artificial intelligence » Attention
