INTRODUCTION: Lung cancer remains a major health challenge, and accurate diagnosis is crucial for effective treatment. Rapid on-site evaluation (ROSE) has emerged as a valuable tool in lung cancer diagnosis. This study aimed to assess the performance of ROSE in the intraoperative diagnosis of lung cancer and its correlation with the histological findings of frozen sections. METHODS: A retrospective review was conducted on 414 cases of intraoperative ROSE performed during lung surgeries between 2017 and 2022. Cytological findings were classified according to the World Health Organization Reporting System for Lung Cytopathology and compared with the subsequent histological diagnoses. RESULTS: ROSE demonstrated a high diagnostic accuracy of 92.2%. The risk of malignancy varied across diagnostic categories, highlighting the value of ROSE in risk stratification. Only 12% of cases with a ROSE diagnosis of malignancy or suspicion of malignancy were discordant with the final histological diagnosis and only as to the specific tumor type. CONCLUSIONS: ROSE is a reliable technique for lung cancer diagnosis, not only during EBUS- or CT-guided FNA but also in the setting of lung surgery, offering high accuracy and risk stratification. It contributes to improved patient management by optimizing procedures, reducing complications, and enhancing specimen quality for further analyses. ROSE is a valuable asset in the modern era of personalized medicine, facilitating tailored treatment approaches.