DataVXAdvanced SciencePaper
journal article·2026-07-20Open access

Multiscale Spatial Fusion Feature‐Driven Characterization of Gastric Cancer Invasive Margins: A Multicenter Cohort Study for Preoperative Accurate Differentiation Between T4a and T4b Subtypes

Guoliang Zheng, Xiaomiao Chai, Peng Jin, Xin Xin, Yingjie Li, Chun Yang, Jia Wei, Zishuo Yan, Jingyu Zhang, Qianning Zhao, Yingchen Han, Ning Zhang, Fuze Li, Bo Qiao, Huan Wang, Huachuan Zheng, Yan Li, Xin Zhang, Yan Zhao, Wenjun Mao, Jing Zhang

Advanced Science
Publisher pageOpen full text
Source abstract

Abstract

ABSTRACT Accurate preoperative differentiation of gastric cancer T4a/b stages is crucial for surgical planning and prognosis. However, conventional CT assessments often yield suboptimal staging accuracy due to visual limitations and inadequate peritumoral microinvasion quantification. This study developed a multi‐scale spatial feature fusion model based on extended regions of interest (eROI) for precise preoperative T4a/b differentiation. We proposed the GAVR model with a three‐tier architecture: a Boundary‐Augmented U‐Net for eROI generation incorporating the peritumoral microenvironment; parallel pathways extracting conventional radiomics, 2D, and 3D deep learning features; and a Vision Transformer for global attention‐weighted fusion and discriminative representation learning. The model was validated across a multicenter cohort of 1804 patients, including internal, external, and prospective sets. A blinded reader study involving 16 radiologists evaluated its clinical utility. GAVR demonstrated exceptional generalizability, achieving AUCs of 0.987 and 0.979 in two independent external sets and 0.987 prospectively. Ablation studies confirmed the necessity of multi‐scale features. GAVR assistance significantly improved radiologists' diagnostic accuracy (0.609 to 0.795) and reduced reading time by 60%. By deeply fusing multi‐scale spatial features, GAVR characterizes structural heterogeneity in complex gastric cancer invasive margins and mitigates overfitting. It demonstrates clear translational value as an embeddable decision‐support tool for multidisciplinary gastric cancer management.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Radiomics and Machine Learning in Medical Imaging · Gastric Cancer Management and Outcomes · Colorectal Cancer Surgical Treatments

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Department of Gastric Surgery Cancer Hospital of Dalian University of Technology Liaoning Cancer Hospital & Institute Shenyang Liaoning China

School of Pharmacy China Medical University Shenyang Liaoning China

Department of Gastric Surgery National Clinical Research Center for Cancer Tianjin Key Laboratory of Digestive Cancer Tianjin Medical University Cancer Institute and Hospital Tianjin Clinical Research Center for Cancer Tianjin China

Shenyang Mental Health Center Shenyang Liaoning China

Department of Radiology Hunnan Central Hospital Shenyang Liaoning China

Department of Medical Oncology The First Hospital of China Medical University Shenyang Liaoning China

School of Management Hebei University Baoding Hebei China

Department of Pathology Cancer Hospital of Dalian University of Technology Liaoning Cancer Hospital & Institute Shenyang Liaoning China

Department of Radiotherapy Cancer Hospital of China Medical University Liaoning Cancer Hospital & Institute Cancer Hospital of Dalian University of Technology Shenyang Liaoning China

Center of Translational Medicine and Department of Gastroenterology The First Affiliated Hospital of Jinzhou Medical University Jinzhou Liaoning China

Department of Laboratory Medicine Institute of Laboratory Medicine Sichuan Provincial People's Hospital School of Medicine University of Electronic Science and Technology of China Chengdu Sichuan China

Department of Nuclear Medicine Shengjing Hospital of China Medical University Shenyang Liaoning China

Department of Thoracic Surgery The Affiliated Wuxi People's Hospital of Nanjing Medical University Wuxi Medical Center Nanjing Medical University Wuxi Jiangsu China

References (30)

Show source-provided references
1. International Agency for Research on Cancer Global Cancer Observatory: Cancer Today – Stomach Cancer Fact Sheet (GLOBOCAN 2022) IARC Lyon France 2024 IARC accessed: December 19 2025 https://gco.iarc.who.int/media/globocan/factsheets/cancers/7‐stomach‐fact‐sheet.pdf.
2. 10.1002/jcla.2355510.1002/jcla.23555 ↗
3. 10.1159/00035088110.1159/000350881 ↗
4. 10.1007/978-3-319-40618-310.1007/978-3-319-40618-3 ↗
5. 10.6004/jnccn.2025.002210.6004/jnccn.2025.0022 ↗
6. 10.1001/2013.jamasurg.30910.1001/2013.jamasurg.309 ↗
7. Japanese Gastric Cancer Association Japanese Gastric Cancer Treatment Guidelines 2014 (ver. 4) Gastric Cancer20(2017):1–19 https://doi.org/10.1007/s10120‐016‐0622‐4.10.1007/s10120-016-0622-4 ↗
8. 10.1007/s10120‐011‐0074‐910.1007/s10120‐011‐0074‐9 ↗
9. 10.1002/bjs.1172610.1002/bjs.11726 ↗
10. 10.1371/journal.pone.010706110.1371/journal.pone.0107061 ↗
11. 10.1002/cac2.1219310.1002/cac2.12193 ↗
12. 10.3748/wjg.v20.i16.454610.3748/wjg.v20.i16.4546 ↗
13. 10.1148/radiol.237304138010.1148/radiol.2373041380 ↗
14. 10.1148/rg.24018610.1148/rg.240186 ↗
15. 10.1007/s11605‐017‐3559‐y10.1007/s11605‐017‐3559‐y ↗
16. 10.21037/qims‐22‐28610.21037/qims‐22‐286 ↗
17. 10.1097/MD.000000000002994710.1097/MD.0000000000029947 ↗
18. 10.12865/CHSJ.48.01.1310.12865/CHSJ.48.01.13 ↗
19. 10.1148/ryai.202019004310.1148/ryai.2020190043 ↗
20. 10.1007/s44178‐022‐00004‐x10.1007/s44178‐022‐00004‐x ↗
21. 10.2196/5685110.2196/56851 ↗
22. 10.1148/radiol.201717192010.1148/radiol.2017171920 ↗
23. China Anti‐Cancer Association Gastric Cancer Committee CACA Guidelines for Holistic Integrative Management of Gastric Cancer Holistic Integrative Oncology1(2022):3 https://doi.org/10.1007/s44178‐022‐00004‐x.10.1007/s44178-022-00004-x ↗
24. 10.1007/s00330‐018‐5887‐410.1007/s00330‐018‐5887‐4 ↗
25. 10.1145/350524410.1145/3505244 ↗
26. 10.1007/s00261‐018‐1547‐210.1007/s00261‐018‐1547‐2 ↗
27. 10.1016/j.media.2021.10228710.1016/j.media.2021.102287 ↗
28. 10.3389/fonc.2022.96341210.3389/fonc.2022.963412 ↗
29. 10.1016/j.acra.2024.12.02910.1016/j.acra.2024.12.029 ↗
30. 10.21147/j.issn.1000‐9604.2018.02.0910.21147/j.issn.1000‐9604.2018.02.09 ↗