Work

Single-inspiratory quantitative CT nomogram for enhanced PRISm and COPD differentiation: a cross-sectional study with interpretable diagnostic boundaries

Year: 2026

Type: article

Abstract: Background Differentiating preserved ratio impaired spirometry (PRISm) from chronic obstructive pulmonary disease (COPD) is challenging. Traditional biphasic CT scans are limited by radiation exposure, while single-inspiratory CT-based deep learning lacks interpretability. This study aimed to develop a single-inspiratory quantitative computed tomography (QCT) nomogram integrating parenchymal, airway, and vascular parameters to redefine imaging definition boundaries. Methods This retrospective study (approved by Ethics Committee YJ-NBEY-KY-2023-107-01) screened 1,265 patients from Ningbo No. 2 Hospital (January 2021–December 2023), yielding 658 eligible participants (Normal: 135, PRISm: 328, COPD: 195) based on predefined inclusion/exclusion criteria. Single-inspiratory CT metrics (parenchymal, airway, vascular) were quantified using the Aview® system. Four logistic regression models distinguished PRISm from normal and COPD group receiver operating characteristic-area under the curve (ROC-AUC) evaluated performance. Results Progressive deterioration in age (COPD: 73.3 vs . PRISm: 69.1 vs . Normal: 64.1 years), male predominance (84.6% COPD vs . 57.9% PRISm), pulmonary function (FEV1%, FEV1/FVC), and CT markers (Pi10: PRISm 3.65 vs . Normal 3.26, P < 0.001) were observed. PRISm showed reduced superficial vessel diameter (AVD9: 2.64 mm vs . Normal 2.95 mm, P < 0.001). Diagnostic models achieved AUCs up to 0.984 (PRISm vs . severe COPD) and 0.853 (PRISm vs . all COPD). Conclusion The QCT nomogram robustly differentiates PRISm from COPD, highlighting reduced superficial vessel diameter as a key biomarker. This radiation-efficient approach enables early COPD stratification via interpretable structural-functional metrics.

Source: PeerJ

DOI: 10.7717/peerj.20575