Background/Objectives: Rapid on-site evaluation (ROSE) is a valuable technique in interventional procedures to immediately assess the adequacy and quality of biopsy specimens at the time they are obtained. The integration of artificial intelligence (AI) into ROSE workflows has demonstrated diagnostic accuracy comparable to that of experienced cytologists. However, clinical implementation of AI-based ROSE models is limited by complex and expensive development. In contrast, the use of free or near-free global large language models (LLMs) offers a significant advantage, making diagnostic support more accessible. Assess the diagnostic accuracy of the LLMs ChatGPT and Gemini in evaluating cytological smears during interventional pulmonology procedures. Methods: Retrospective evaluation of the efficacy of LLMs for assessment of cytological smears obtained from adult patients who underwent interventional bronchoscopic and ultrasound-guided biopsies between 2020 and 2025. Images of ROSE-prepared samples were analyzed by ChatGPT-4o, ChatGPT-5, ChatGPT-5 “thinking”, and Gemini 2.5 models. Results: Forty-eight procedures in 47 patients (mean age 65 years) were analyzed; 79% of biopsies were malignant. Using the final histopathology report as reference, cytologists achieved balanced accuracy of 0.75 (Gwet’s AC1 = 0.53, sensitivity 0.71, specificity 0.78). ChatGPT-5 “thinking” showed high concordance (accuracy 0.65, Gwet’s AC1 = 0.81, sensitivity 1.00, specificity 0.30). Gemini reached an accuracy of 0.59 (Gwet’s AC1 = 0.76, sensitivity 0.97, specificity 0.20). Conclusions: To our knowledge, this study is the first to evaluate LLM-assisted ROSE in interventional pulmonology. The results suggest the potential feasibility of integrating this AI technology into the workflow within the pulmonary division. Larger prospective studies are needed to confirm effects on diagnostic yield.

Background: Computed tomography (CT)-guided lung biopsy plays a pivotal role in diagnosing thoracic lesions. However, its diagnostic yield may be compromised in large, necrotic, or heterogeneous tumours due to inadvertent sampling of non-viable tissue. Dual-energy CT (DECT) iodine mapping provides functional imaging by identifying iodine-avid, perfused areas, thereby offering the potential to improve biopsy targeting. Methods: This single-centre retrospective study evaluated the clinical feasibility and diagnostic performance of DECT-guided biopsy. Adult patients with suspected necrotic lung or mediastinal lesions who underwent DECT iodine mapping prior to CT-guided biopsy between April 2021 and December 2022 were evaluated. DECT iodine maps were generated using dual-source CT and used to identify viable tumour regions for targeted biopsy. The primary outcome was diagnostic yield, defined as obtaining a definitive histopathological diagnosis. Secondary outcomes included safety and adequacy of samples for molecular testing. Results: Twenty patients were included. A definitive diagnosis was obtained in 18/20 biopsies (90%). Diagnostic yield was 9/11 (81.8%) for pulmonary lesions and 9/9 (100%) for mediastinal/pleural lesions. Diagnoses included non-small-cell lung cancer (n = 8), Hodgkin lymphoma (n = 4), thymoma (n = 3), and other malignancies (n = 3). Biopsy material was sufficient for additional molecular testing in 13/20 cases (65%). Complications were minor (one pneumothorax not requiring drainage and two self-limited bleeding events). Conclusions: DECT iodine map-guided targeting was feasible in this retrospective cohort and was associated with high diagnostic yield, low complication rates, and frequent acquisition of tissue suitable for molecular analyses. Prospective controlled studies are needed to quantify benefit over conventional CT guidance.