Rapid On-Site AI-Assisted Grading for Lung Surgery Based on Optical Coherence Tomography

SIMPLE SUMMARY: In early-stage lung cancer surgery, determining the extent of resection relies on microscopic examination of frozen sections (FSs), especially when the histology is unknown preoperatively. While optical coherence tomography (OCT) holds promise for instant lung cancer diagnosis, grading tumors with OCT remains challenging. Our study proposes an interactive human–machine interface (HMI) that integrates a mobile OCT system, deep learning, and attention mechanisms. The interactive HMI can mark lesion locations on real-time images and perform tumor grading, aiding clinical decisions. In a trial with twelve preoperatively indeterminate adenocarcinoma patients who underwent thoracoscopic resection, the results of the presented HMI system outperformed frozen sections, achieving an 84.9% overall accuracy compared to FSs’ 20%, showcasing the HMI’s potential for rapid diagnostics and improved patient outcomes. ABSTRACT: The determination of resection extent traditionally relies on the microscopic invasiveness of frozen sections (FSs) and is crucial for surgery of early lung cancer with preoperatively unknown histology. While previous research has shown the value of optical coherence tomography (OCT) for instant lung cancer diagnosis, tumor grading through OCT remains challenging. Therefore, this study proposes an interactive human–machine interface (HMI) that integrates a mobile OCT system, deep learning algorithms, and attention mechanisms. The system is designed to mark the lesion’s location on the image smartly and perform tumor grading in real time, potentially facilitating clinical decision making. Twelve patients with a preoperatively unknown tumor but a final diagnosis of adenocarcinoma underwent thoracoscopic resection, and the artificial intelligence (AI)-designed system mentioned above was used to measure fresh specimens. Results were compared to FSs benchmarked on permanent pathologic reports. Current results show better differentiating power among minimally invasive adenocarcinoma (MIA), invasive adenocarcinoma (IA), and normal tissue, with an overall accuracy of 84.9%, compared to 20% for FSs. Additionally, the sensitivity and specificity, the sensitivity and specificity were 89% and 82.7% for MIA and 94% and 80.6% for IA, respectively. The results suggest that this AI system can potentially produce rapid and efficient diagnoses and ultimately improve patient outcomes.