AI-powered hierarchical classification of ampullary neoplasms: a deep learning approach using white-light and narrow-band imaging.

Journal: Surgical endoscopy
Published Date: Jan 14, 2026

Abstract

BACKGROUND: Endoscopic diagnosis of Ampulla of Vater (AoV) lesions remains challenging owing to complex morphology and limited representative images, particularly for high-risk dysplastic lesions. This study aimed to develop a hierarchical deep learning framework for the stepwise classification of ampullary lesions using white-light (WL) and narrow-band endoscopic images (NBI). METHODS: The framework employs three sequential binary classifications: (1) normal vs. abnormal, (2) adenoma vs. cancer, and (3) high-grade dysplasia (HGD) vs. low-grade dysplasia (LGD) within adenomas. Each stage uses EfficientNet-B4 classifiers trained independently on WL and NBI. Predictions are integrated using confidence-based voting. To overcome data scarcity and class imbalance, for HGD and cancer, we used StyleGAN2-ADA to generate synthetic images. The hierarchical model was developed using 4244 endoscopic images from 464 patients collected at Seoul National University Hospital (2693/833/718 for train/validation/test). RESULTS: The hierarchical model achieved stage-specific accuracies of 95.6% (normal vs. abnormal), 94.4% (adenoma vs. cancer), and 92.7% (LGD vs. HGD), resulting in overall diagnostic accuracy of 92.2%. The model demonstrated excellent sensitivity of 83.3% for HGD and 87.5% for cancer, with specificities exceeding 98%. The confidence-based dual-modality approach (AUROC: 0.921) significantly outperformed single-modality approaches using WL alone (AUROC: 0.866) or NBI alone (AUROC: 0.895), by integrating their complementary diagnostic strengths. Generative adversarial network-based augmentation substantially improved sensitivity for cancer (from 87.5% to 91.7%) and HGD (from 83.3% to 86.5%), while overall accuracy increased from 94.5% to 95.1%. CONCLUSIONS: A hierarchical deep learning approach integrating dual-modality imaging and synthetic data augmentation significantly improves diagnostic performance for ampullary lesions.

Authors

  • Dan Yoon
    Interdisciplinary Program in Bioengineering, Graduate School, Seoul National University, Seoul, 08826, South Korea.
  • Sung Hoon Chang
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul National University Hospital, Seoul, 03080, Republic of Korea.
  • Woo Hyun Paik
    Department of Internal Medicine and Liver Research Institute, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea.
  • Chang Hyun Kim
    Department of Surgery, Chonnam National University Hwasun Hospital, Chonnam National University Medical School, Hwasun, Korea.
  • Byeong Soo Kim
    Interdisciplinary Program in Bioengineering, Graduate School, Seoul National University, Seoul, Republic of Korea.
  • Young Gyun Kim
    Interdisciplinary Program in Bioengineering, Seoul National University Graduate School, Seoul, Republic of Korea.
  • Hyunsoo Chung
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul, 03080, Republic of Korea. [email protected].
  • Ji Kon Ryu
    Department of Internal Medicine and Liver Research Institute, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea.
  • Sang Hyub Lee
    Department of Urology, Kyung Hee University School of Medicine, Seoul, Korea.
  • In Rae Cho
    Department of Internal Medicine and Liver Research Institute, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea.
  • Seong Ji Choi
  • Joo Seong Kim
    Department of Internal Medicine, Dongguk University College of Medicine, Dongguk University Ilsan Hospital, Goyang, Korea.
  • Sungwan Kim
    Department of Biomedical Engineering, Seoul National University College of Medicine, Seoul, Republic of Korea.
  • Jin Ho Choi
    Department of Medical Device Management and Research, SAIHST, Sungkyunkwan University, Seoul 06351, South Korea; Department of Emergency Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul 06351, South Korea. Electronic address: [email protected].

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals