Identification of Pulmonary Embolism in Patients With Chronic Obstructive Pulmonary Disease Using Clinical and Non-contrast CT Radiomic Features
Year: 2026
Type: preprint
Study design: Observational Study
Abstract: Pulmonary embolism (PE) is a potentially life-threatening complication in patients with chronic obstructive pulmonary disease (COPD), but early recognition is challenging because of substantial overlap in clinical manifestations. We aimed to develop and internally validate a multimodal model integrating clinical variables and radiomic features derived from non-contrast chest computed tomography (CT) to identify concomitant PE in patients with COPD. In this retrospective study, 415 patients with COPD were included, of whom 116 had concomitant PE confirmed by computed tomography pulmonary angiography (CTPA). Radiomic features were extracted from non-contrast chest CT images. Feature selection was performed using a multistep procedure incorporating univariable screening, correlation-based redundancy reduction, minimum redundancy maximum relevance, and least absolute shrinkage and selection operator regression. Clinical, radiomics, and combined models were developed and internally validated. Model performance was assessed using receiver operating characteristic curve analysis, calibration analysis, and decision curve analysis. The combined model showed the highest discriminative performance, with an area under the receiver operating characteristic curve of 0.987 in the internal test cohort, compared with 0.959 for the clinical model and 0.937 for the radiomics model. The combined model also demonstrated good calibration and yielded greater net clinical benefit across a broad range of threshold probabilities. A multimodal model integrating routinely available clinical variables with radiomic features from non-contrast chest CT showed excellent performance for identifying concomitant PE in patients with COPD. This approach may enable opportunistic PE risk stratification during routine chest CT examinations and help identify patients who may benefit from further CTPA evaluation. External validation is warranted before clinical implementation.
Source: Research Square
Institution: Zhejiang University
Language: English
Subfield: Radiology, Nuclear Medicine and Imaging
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7214001968