Clinical, Physiological, Immunological, and Radiological Dataset for the Derivation and Internal Validation of the NRS-ILD Score in Diffuse Parenchymal Lung Disease

This dataset contains prospective, de-identified clinical, physiological, immunological, and volumetric high-resolution computed tomography (HRCT) data from 50 adult patients with confirmed diffuse parenchymal lung disease (DPLD) evaluated at Nilratan Sircar Medical College and Hospital, Kolkata, India. The dataset includes demographic profiles, anthropometry, resting spirometry, hemoglobin-adjusted single-breath diffusing capacity of the lungs for carbon monoxide (DLCO), 6-minute walk test (6MWT) dynamic exertion metrics, modified Borg dyspnea scale scores, mMRC dyspnea grading, anti-nuclear antibody (ANA) seropositivity, and HRCT lobar involvement patterns. This repository serves as the primary data source for the multivariable derivation, Harrell's optimism-corrected cross-validation, and bootstrap internal validation (1,000 resamples) of the NRS-ILD Comprehensive Index and NRS-ILD Bedside Score.