Development and Crossover Evaluation of an Artificial Intelligence-Assisted System for Solid Pancreatic Lesion Detection and Pancreatic Parenchyma Recognition in Endoscopic Ultrasonography (With Video)
In brief
AI boosts novice EUS lesion detection to 89% (vs 77%)
In a six-center crossover study, an AI overlay raised novice endosonographers' sensitivity for solid pancreatic lesions from 77% to nearly 89% without sacrificing specificity. Accuracy also improved, while experts maintained performance and gained modest specificity gains. The tool shows promise as an adjunct for training, but larger trials are needed to confirm clinical impact.
- Journal
- Digestive endoscopy : official journal of the Japan Gastroenterological Endoscopy Society (Q1)
- Published
- 1 September 2026
- Study design
- Randomized controlled trial
- Evidence level
- Level 1, High (CEBM 1b)
- Authors
- Toshio Fujisawa, Takamichi Kuwahara, Tatsuya Sato, Yusuke Takasaki, Nozomi Okuno, Ryunosuke Hakuta, et al.
- PMID
- 42720264
- DOI
- 10.1111/den.70282
Why clinicians should know about it
- Picked for Radiology, Radiation Oncology, Nuclear Medicine, Medical Physics and Imaging (top studies of the week, 13 September 2026): Phase III RCT, practice-changing for adjuvant therapy
- Picked for Gastroenterology (top studies of the week, 13 September 2026): AI assistance improves solid pancreatic lesion detection
Abstract
BACKGROUND AND STUDY AIMS: Pancreatobiliary endoscopic ultrasonography (EUS) is technically demanding, and supervised training opportunities are limited. We developed an artificial intelligence (AI) overlay system for detecting solid pancreatic lesions (SPL) and recognizing pancreatic parenchyma (PP) and evaluated its effect on reader performance. PATIENTS AND METHODS: Across six centers, two deep learning-based models were trained using expert-annotated EUS frames. We then conducted a randomized, two-sequence, two-period crossover reader study in which eight endosonographers (five novices and three experts) interpreted image sets with and without AI assistance. The primary endpoint was superiority of sensitivity for SPL detection among novices; key secondary endpoints included specificity and PP recognition. RESULTS: From 118 patients, 120 SPL-positive/negative image sets and 160 PP-positive/negative image sets were constructed. Among novices, AI assistance improved SPL detection sensitivity (88.7% vs. 76.8%, p < 0.001) and accuracy (86.4% vs. 78.7%), while specificity met the predefined noninferiority criterion (84.2% vs. 80.5%, p < 0.001). For PP recognition, sensitivity increased numerically (86.3% vs. 83.3%) but did not meet the predefined superiority criterion (p = 0.095); specificity met the noninferiority criterion (87.8% vs. 81.0%), and accuracy increased from 82.1% to 87.0%. Among experts, sensitivity was maintained for both tasks, whereas specificity increased with AI assistance. CONCLUSIONS: AI assistance improved SPL detection among novice endosonographers. For PP recognition, sensitivity increased without reaching statistical superiority, whereas specificity met the predefined noninferiority criterion. These findings support a potential adjunctive role for AI in EUS interpretation.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.