Dual geometry-inspired structural modeling for neuroendocrine tumor segmentation in endoscopic ultrasound
- 1 July 2026
- journal article
- Published by Springer Nature in International Journal of Computer Assisted Radiology and Surgery
- Vol. 21 (7) , 1631-1638
Abstract
The increasing incidence of lower gastrointestinal neuroendocrine tumors (NETs) necessitates improved methods for early and accurate detection. Automatic segmentation of NETs in endoscopic ultrasound (EUS) images is particularly challenging due to low image contrast and indistinct tumor boundaries. This study proposes and validates GismEUS, a geometry-aware deep learning model for automated NET segmentation in EUS images. We propose GismEUS, a deep learning architecture that integrates dual features. The model employs Endoscopic Planar Geometric Feature Projection to capture fine-grained local features and Endoscopic Stereoscopic Structural Feature Modeling to extract comprehensive global features. These feature sets are fused by the Dual Structure-Guided Feature Enhancement module, which applies both semantic-aware and distance-aware attention and subsequently combines their outputs via a local-global gated fusion. The model was trained and evaluated on a private annotated dataset for EUS images of NETs and the public GIST514-DB dataset. GismEUS demonstrated superior segmentation performance across multiple evaluation metrics, achieving a Dice score of 0.6735, and significantly outperformed established benchmarks on the EUS datasets. These comprehensive results validate the effectiveness of the proposed feature integration strategy in addressing the unique challenges of EUS imaging. This study demonstrates that the novel combination of planar and stereoscopic features significantly improves segmentation accuracy and introduces a high-quality annotated dataset for NETs segmentation in EUS images. GismEUS model offers a reliable and accurate automated tool for early NETs diagnosis, supporting more consistent clinical decision-making and potentially improving patient outcomes.Keywords
Funding Information
- National Natural Science Foundation of China (82572314)
- Key Technologies Research and Development Program (2025YFC2426300)
- Key Technologies Research and Development Program (62271246)
- Science and Technology Commission of Shanghai Municipality (24511104100)
This publication has 19 references indexed in Scilit:
- A self-supervised framework for improved generalisability in ultrasound B-mode image segmentationBiomedical Signal Processing and Control, 2025
- A dual-branch network for ultrasound image segmentationBiomedical Signal Processing and Control, 2025
- Bimodal segmentation and classification of endoscopic ultrasonography images for solid pancreatic tumorBiomedical Signal Processing and Control, 2023
- Query: Query over queries for improving gastrointestinal stromal tumour detection in an endoscopic ultrasoundComputers in Biology and Medicine, 2022
- Asymmetric U-shaped network with hybrid attention mechanism for kidney ultrasound images segmentationExpert Systems with Applications, 2022
- Epidemiologic Trends of and Factors Associated With Overall Survival for Patients With Gastroenteropancreatic Neuroendocrine Tumors in the United StatesJAMA Network Open, 2021
- Deep learning–based pancreas segmentation and station recognition system in EUS: development and validation of a useful training tool (with video)Gastrointestinal Endoscopy, 2020
- UNet++: A Nested U-Net Architecture for Medical Image Segmentation2018
- Clinical impact of endoscopic ultrasonography on the management of neuroendocrine tumors: lights and shadowsDigestive and Liver Disease, 2018
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015