Pulmonary Embolism Risk Stratification from CTPA and Medical Records: Vascular Graphs Are Not All You Need

2026-06-24

AI Summary

This study addresses the challenge of accurately stratifying risk in patients with pulmonary embolism using only CT pulmonary angiography (CTPA) images and clinical records, without access to blood test data. The authors propose a unified framework that integrates clinical notes, cardiac biomarkers, and structural information from pulmonary vascular graphs, combining tabular models with graph neural networks (GNNs). For the first time, they systematically evaluate the discriminative value of vascular graph structures for risk stratification. Experiments on a cohort of 353 complete cases demonstrate that a tabular model leveraging only clinical records and biomarkers outperforms GNN-based approaches incorporating vascular graph data, suggesting that vascular graph structure provides no significant added benefit for risk stratification—a finding that challenges prevailing assumptions in the field that rely on vascular morphology modeling.

Abstract

Risk stratification for pulmonary embolism (PE) is critical for clinical decision-making. Stratification guidelines are based on patient medical records, parameters measured from computed tomography pulmonary angiography (CTPA), and blood tests. However, blood tests are often missing in routine practice. This work studies whether state-of-the-art models can accurately classify risk stratification from only medical records and biomarkers extracted from CTPA images. We benchmark different approaches to combine medical records and cardiac biomarkers with rich pulmonary vascular information; we add vascular biomarkers to tabular models and apply graph neural networks (GNNs) on the vascular tree's intrinsic graph representation. We use a private dataset (n=353) with uniquely complete data for PE risk stratification. Our results show that, among global features, medical records and cardiac biomarkers are the most significant predictors, while vascular biomarkers do not further improve stratification. Even more surprising, even GNNs on vascular graphs fail to outperform strong tabular baseline on global features. We consider hypotheses, on both models and data, that could explain this suboptimal performance. Our investigation suggests that, counter-intuitively, vascular graphs might hold no discriminative information for PE risk stratification. Code is available from https://github.com/creatis-myriad/GENESIS.

Problem

• Pulmonary Embolism
• Risk Stratification
• CTPA
• Vascular Graphs
• Medical Records

Innovation

• pulmonary embolism
• risk stratification
• graph neural networks
• vascular biomarkers
• CTPA