Abstract. The aim of the present study was to compare pathological diagnoses, as determined by the new International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society (IASLC/ATS/ERS) classification, with conventional radiological features. In addition, the present study aimed to evaluate the correlation among clinical characteristics, computed tomography (CT) images and gene mutation status in patients with stage IA adenocarcinoma of the lung. A total of 212 patients with stage IA lung adenocarcinoma were included in the study. The patients were classified into pure ground-glass opacity (pGGO), mixed GGO (mGGO) and solid GGO (sGGO) by CT imaging. Histological subtype was classified according to the IASLC/ATS/ERS classification of lung adenocarcinoma. In addition, epidermal growth factor receptor (EGFR) and Kirsten rat sarcoma (KRAS) mutation assays were performed, and 36.8% of patients (78/212) were determined to have an EGFR mutation, while 8.5% of patients (18/212) were found to have a KRAS mutation. According to the IASLC/ATS/ERS classification, 44 cases were diagnosed as adenocarcinoma in situ (AIS; 20.8%), 62 cases were diagnosed as minimally invasive adenocarcinoma (MIA; 29.2%) and 106 cases were classified as invasive adenocarcinoma (IAC; 50.0%). pGGO image patterns were observed in 39.2% of patients (n=83), while mGGO and sGGO patterns were observed in 28.8% (n=61) and 32.0% (n=68) of patients, respectively. From pGGO to sGGO, cases of AIS and MIA were shown to have a decreasing trend, while IAC cases exhibited an increasing trend (P=0.036). Analysis of the correlation between CT image patterns and gene mutations demonstrated that L858R point mutations, exon 19 deletions and KRAS mutations were more common in lesions with a lower GGO proportion (P=0.029, 0.027 and 0.018, respectively). Therefore, according to the IASLC/ATS/ERS classification, GGO imaging patterns were shown to correlate with subtypes of adenocarcinomas. In addition, EGFR and KRAS mutations were found to be associated with lesions with a low GGO proportion. Therefore, analysis of GGO lesions may offer useful indications of the histological subtype of an adenocarcinoma in patients with stage IA lung adenocarcinoma, and predictive value for EGFR and KRAS mutations.