Introduction
Pancreatic neuroendocrine neoplasms (PNENs) are rare tumors with heterogeneous outcomes. Tumor grading (G), based on mitotic count and Ki-67 index, is the main prognostic factor guiding treatment. Endoscopic ultrasound-guided fine-needle aspiration/biopsy (EUS-FNA/B) is the current standard but shows a misgrading rate up to 25%.
Aims & Methods
We evaluated an artificial intelligence (AI)–based video analysis model to predict PNEN grading from CE-EUS recordings. This retrospective study was conducted at IRCCS San Raffaele Hospital, Milan, an ENETS Center of Excellence. Patients were eligible if CE-EUS videos >=1 minute (arterial and venous phases) and cyto-histological confirmation of PNEN were available. Exclusion criteria included mixed neuroendocrine–non-neuroendocrine neoplasms, missing Ki-67 grading, or poor video quality. CE-EUS videos were processed with a deep-learning video transformer model (GradAINet). The dataset was split into training (70%), validation (10%), and testing (20%) cohorts. Diagnostic performance was evaluated using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and F1-score.
Results
Between 2022 and 2024, 115 patients were included (49 female, 42.6%): 70 had G1 and 45 had G2–G3 tumors. Overall, 253,751 video frames were analyzed. GradAINet achieved a sensitivity of 0.817 (95% CI: 0.556–1.000), specificity 0.806 (95% CI: 0.588–1.000), PPV 0.759 (95% CI: 0.500–1.000), NPV 0.856 (95% CI: 0.667–1.000), and accuracy 0.811 (95% CI: 0.654–0.962).
Conclusion
This AI-driven CE-EUS video model shows high accuracy for PNEN grading and potentially complementing EUS-FNA/B. As the first video-based rather than static-image model, it represents a methodological advance. Multicenter validation on larger cohorts is needed before clinical implementation.