Deep learning model for pathological invasiveness prediction using smartphone-based surgical resection images in clinical stage IA lung adenocarcinoma (SuRImage): a prospective, multicentric, diagnost
5 hours ago
- #lung adenocarcinoma
- #surgical decision support
- #deep learning
- A deep learning model named SuRImage uses smartphone-captured surgical resection images to predict pathological invasiveness in clinical stage IA lung adenocarcinoma, aiding intraoperative decision-making.
- The model achieved high diagnostic performance with AUCs of 0.84 for invasive identification, 0.87 for diagnosis, and 0.85 for grading, outperforming frozen section analysis and improving surgeon accuracy.
- Conducted as a prospective multicentric study in China, it enrolled patients from three hospitals and highlights the potential for streamlined surgical workflows based on macroscopic morphological features.