Deep learning model targeting cancer - surrounding tissues for accurate cancer diagnosis based on histopathological images

Sep 23, 2026

Uploaded by Jiven

Accurate and fast histological cancer diagnosis is essential for effective treatment. Deep learning methods have supported pathologists, and remodeled microenvironments and field cancerization suggest cancer-specific signals may exist in non-cancer regions around tumors. The study proposes a fine-tuned framework for gastric cancer that targets regions of interest (ROI) with varying proportions of no-cancer and cancer areas, and evaluates how diagnostic performance depends on that proportion. An optimized MobileNetV2 model (DeepNCCNet) achieved 93.96% accuracy in external validation, highlighting the clinical value of non-cancer information when tumor tissue is limited in biopsy.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Abstract

Highlights

Keywords

Introduction

Li et al. Journal of Translational Medicine (2024) 23:110 [https://doi.org/10.1186/s12967-024-06017-6](https://doi.org/10.1186/s12967-024-06017-6) Journal of Translational Medicine RESEARCH Open Access Deep learning model targeting cancer surrounding tissues for accurate cancer diagnosis based on histopathological images Lanlan Li1†, Yi Geng1†, Tao Chen1†, Kaixin Lin2,3,4,5,6†, Chengjie Xie1, Jing Qi1, Hongan Wei1, Jianping Wang7, Dabiao Wang8, Ze Yuan3,4,5, Zixiao Wan3,4,5, Tuoyang Li2,3,4,5, Yanxin Luo2,3,4,5,6, Decao Niu9*, Juan Li2,4,5,10* and Huichuan Yu2,3,4,5,6* Abstract Accurate and fast histological diagnosis of cancers is crucial for successful treatment. The deep learning-based approaches have assisted pathologists in efficient cancer diagnosis. The remodeled microenvironment and field cancerization may enable the cancer-specific features in the image of non-cancer regions surrounding cancer, which may provide additional information not available in the cancer region to improve cancer diagnosis. Here, we proposed a deep learning framework with fine-tuning target proportion towards cancer surrounding tissues in histological images for gastric cancer diagnosis. Through employing six deep learning-based models targeting region-of-interest (ROI) with different proportions of no-cancer and cancer regions, we uncovered the diagnostic value of non-cancer ROI, and the model performance for cancer diagnosis depended on the proportion. Then, we constructed a model based on MobileNetV2 with the optimized weights targeting non-cancer and cancer ROI to diagnose gastric cancer (DeepNCCNet) . In the external validation, the optimized DeepNCCNet demonstrated excellent generalization abilities with an accuracy of 93. 96% . In conclusion, we discovered a non-cancer ROI weight-dependent model performance, indicating the diagnostic value of non-cancer regions with potential remodeled microenvironment and field cancerization, which provides a promising image resource for cancer diagnosis. The DeepNCCNet could be readily applied to clinical diagnosis for gastric cancer, which is useful for some clinical settings such as the absence or minimum amount of tumor tissues in the insufficient biopsy. Highlights • A deep learning framework with optimized non-cancer and cancer ROI for accurate cancer diagnosis • The non-cancer region is a promising resource for the current deep-learning framework to improve cancer diagnosis. †Lanlan Li, Yi Geng, Tao Chen, and Kaixin Lin contributed equally to this work. *Correspondence: Decao Niu[ndcdoct@sr.gxmu.edu.cn](ndcdoct@sr.gxmu.edu.cn)[ ](ndcdoct@sr.gxmu.edu.cn)Juan Li[lijuan67@mail.sysu.edu.cn](lijuan67@mail.sysu.edu.cn)[ ](lijuan67@mail.sysu.edu.cn)Huichuan Yu[yuhch5@mail.sysu.edu.cn](yuhch5@mail.sysu.edu.cn) Full list of author information is available at the end of the article © The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit [http://creativecommons.org/licenses/by-nc-nd/4.0/](http://creativecommons.org/licenses/by-nc-nd/4.0/.)[.](http://creat

FAQs

What motivates the use of non-cancer surrounding tissues for cancer diagnosis in this study?

The study argues that remodeled microenvironment and field cancerization may embed cancer-specific features in images from non-cancer regions around tumors, providing additional diagnostic information beyond the cancer region.

How does the proposed framework incorporate the proportions of no-cancer versus cancer regions?

The framework fine-tunes the target proportion by training six deep learning models using ROIs with different balances of no-cancer and cancer areas, then analyzes how model performance changes with that proportion.

What is the role and result of the DeepNCCNet model for gastric cancer diagnosis?

DeepNCCNet is built on MobileNetV2 with optimized weights targeting non-cancer and cancer ROIs. External validation showed excellent generalization performance with an accuracy of 93.96%, supporting clinical applicability when biopsy contains insufficient tumor tissue.

Sep 23, 2026

Uploaded by Jiven