biostudies-literatureKhorrami MBLRD VANCATS NIH HHSNIBIB NIH HHSCongressionally Directed Medical Research ProgramsNCRR NIH HHSNHLBI NIH HHSNational Cancer InstituteNCI NIH HHSDepartment of Veterans Affairs146-158https://www.ebi.ac.uk/biostudies/studies/S-EPMC8087632biostudies-literatureUnknown148<h4>Objective</h4>To identify stable and discriminating radiomic features on non-contrast CT scans to develop more generalisable radiomic classifiers for distinguishing granulomas from adenocarcinomas.<h4>Methods</h4>In total, 412 patients with adenocarcinomas and granulomas from three institutions were retrospectively included. Segmentations of the lung nodules were performed manually by an expert radiologist in a 2D axial view. Radiomic features were extracted from intra- and perinodular regions. A total of 145 patients were used as part of the training set (S_tr), whereas 205 patients were used as part of test set I (S_te¹) and 62 patients were used as part of independent test set II (S_te²). To mitigate the variation of CT acquisitionEuropean journal of cancer (Oxford, England : 1990)Distinguishing granulomas from adenocarcinomas by integrating stable and discriminating radiomic features on non-contrast computed tomography scans.PMC8087632R01 HL151277UL1 TR002548U54 CA254566C06 RR012463R43 EB028736R01 CA216579U01 CA248226R01 CA249992I01 BX004121U24 CA199374U01 CA239055R01 CA220581R01 CA202752R01 CA208236Gupta ALinden PThawani RPennell NKhorrami MRajiah PBera KJacono FGilkeson RCMadabhushi AFu PVelcheti VfalseDistinguishing granulomas from adenocarcinomas by integrating stable and discriminating radiomic features on non-contrast computed tomography scans.<h4>Objective</h4>To identify stable and discriminating radiomic features on non-contrast CT scans to develop more generalisable radiomic classifiers for distinguishing granulomas from adenocarcinomas.<h4>Methods</h4>In total, 412 patients with adenocarcinomas and granulomas from three institutions were retrospectively included. Segmentations of the lung nodules were performed manually by an expert radiologist in a 2D axial view. Radiomic features were extracted from intra- and perinodular regions. A total of 145 patients were used as part of the training set (S_tr), whereas 205 patients were used as part of test set I (S_te¹) and 62 patients were used as part of independent test set II (S_te²). To mitigate the variation of CT acquisition2021-01-01T00:00:00Z2021 May2025-04-04T10:17:57.397Z2025-04-04T10:17:57.397ZS-EPMC80876323374348310.1016/j.ejca.2021.02.008