Xu Y, Lu L, Lin-ning E, et al. - Patients with a single pulmonary nodule (PNs (n = 373) who underwent a preoperative chest CT were enrolled retrospectively at Beijing Friendship Hospital from March 2015 to March 2018 in order to examine the utility of radiomics for prognosticating the malignancy of PNs of various sizes using unenhanced, thin-section CT images. The most contributive radiomic peculiarities for prognosticating PN malignancy for groups T1a, T1b, and T1c were LoG_X_Uniformity, Intensity_Minimum, and Shape_SI9, respectively. Thus, to prognosticate the malignancy of pulmonary nodules, radiomic characteristics on the basis of unenhanced CT images could be used. Better prognostication performance to the T1b and T1c models was exhibited by the radiomic T1a model, and in terms of AUC and sensitivity, the best performance was discovered for prognosticating the malignancy of T1a PN.