Table of Links
1. Introduction
Pulmonary embolism (PE) is a significant cause of morbidity and mortality, with nearly 600,000 cases and 60,000 deaths annually in the United States.[1,2] Following acute myocardial infarction and stroke, PE is the next most common cause of cardiovascular death in hospitalized patients.[3] Efficient diagnosis and management is key, as most deaths (>70%) occur within the first hour.[1] Clinical presentation is highly variable: common symptoms include tachycardia, dyspnea, and pleuritic chest pain.[4]
Given that timely and accurate risk stratification is vital for PE management, several prognostication tools have been developed. The Pulmonary Embolism Severity Index (PESI) is a well-validated tool that estimates 30-day mortality in patients with acute PE based on 11 clinical variables, serving as the present gold standard.[5] While PESI boasts a 99% negative predictive value of deterioration in patients classified as low-risk, positive predictive value for high-risk patients remains suboptimal at 11%.[6] Thus, there is a longstanding need to improve prognostication following diagnosis. Traditional survival methods include random survival forest (RSF), which employs a tree-based ensemble model, and Cox proportional hazards (CoxPH) models, which utilize hazard functions to estimate the linear impact of covariates on risk.[7,8]
With recent advances in artificial intelligence (AI), deep learning (DL)-based approaches have emerged as promising alternatives, significantly augmenting the interpretation of medical imaging studies.[9,10] AI-based models employed on computed tomography pulmonary angiography (CTPA) have been shown to diagnose PE with high accuracy and predict clot burden in acute cases.[11-14] Utilizing multimodal survival data, survival analysis techniques utilizing multimodal learning have exhibited enhanced robustness compared to single-modality techniques.[15,16] We hypothesized that a prognostication model combining imaging and clinical data would outperform PESI alone. We therefore aimed to develop and validate DL models using CTPA and clinical data to predict mortality in patients with PE.
Authors:
(1) Zhusi Zhong, BS, a Co-first authors from Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA, Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA, and School of Electronic Engineering, Xidian University, Xi’an 710071, China;
(2) Helen Zhang, BS, a Co-first authors from Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA and Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA;
(3) Fayez H. Fayad, BA, a Co-first authors from Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA and Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA;
(4) Andrew C. Lancaster, BS, Department of Radiology and Radiological Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 21205, USA and Johns Hopkins University School of Medicine, Baltimore, MD, 21205, USA;
(5) John Sollee, BS, Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA and Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA;
(6) Shreyas Kulkarni, BS, Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA and Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA;
(7) Cheng Ting Lin, MD, Department of Radiology and Radiological Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 21205, USA;
(8) Jie Li, PhD, School of Electronic Engineering, Xidian University, Xi’an 710071, China;
(9) Xinbo Gao, PhD, School of Electronic Engineering, Xidian University, Xi’an 710071, China;
(10) Scott Collins, Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA and Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA;
(11) Colin Greineder, MD, Department of Pharmacology, Medical School, University of Michigan, Ann Arbor, MI, 48109, USA;
(12) Sun H. Ahn, MD, Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA and Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA;
(13) Harrison X. Bai, MD, Department of Radiology and Radiological Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 21205, USA;
(14) Zhicheng Jiao, PhD, Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA and Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA;
(15) Michael K. Atalay, MD, PhD, Department of Diagnostic Radiology, Rhode Island Hospital, Providence, RI, 02903, USA and Warren Alpert Medical School of Brown University, Providence, RI, 02903, USA.
This paper is