{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T12:00:06Z","timestamp":1780574406245,"version":"3.54.1"},"reference-count":25,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T00:00:00Z","timestamp":1780531200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Gastrointestinal precancerous lesions, including colorectal polyps and gastric intraepithelial neoplasia, share highly similar morphological and pathological characteristics. In this study, we validate our lightweight attention-enhanced model on the Kvasir colorectal dataset, which serves as a representative benchmark for gastrointestinal lesion classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Early detection of gastric precancerous lesions is crucial for reducing gastric cancer mortality. This study aimed to address the trade-off between lightweight performance, classification accuracy and interpretability of existing AI models for endoscopic image analysis, and develop a practical attention-enhanced lightweight model for gastrointestinal precancerous lesion classification to support clinical application.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>An attention-enhanced lightweight MobileNetV3 model was proposed by integrating a Spatial-Channel Attention (SCA) module after the last convolutional layer of the backbone for dual spatial and channel weighting to enhance lesion-focused feature extraction. Ablation experiments were conducted to verify the SCA module\u2019s effectiveness by comparing models with and without the module. GradCAM was used to visualize the model\u2019s decision-making process. The model was evaluated on the Kvasir dataset containing 3,000 endoscopic images (3 classes, normal gastrointestinal mucosa, polyp, ulcerative colitis) with a 7:2:1 train-validation-test split. The training adopted Adam optimizer (lr\u202f=\u202f1e\u22124), StepLR scheduler and early stopping (patience\u202f=\u202f5) to ensure stability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The proposed model achieved an overall accuracy of 99.10% and a weighted F1-score of 99.10%. For polyp (the key high-risk precancerous lesion), it reached 100% precision and 98% recall. Ablation experiments confirmed that the SCA module stably improved core metrics by 0.07% without increasing computational redundancy. The model maintained ultra-lightweight characteristics with only 1.05\u202fM total parameters and 3.99 MB parameter storage, supporting real-time deployment on endoscopic devices. GradCAM heatmaps showed that the SCA module enabled the model to precisely focus on clinically relevant regions (e.g., polyp edges), consistent with endoscopists\u2019 observation habits.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The proposed model effectively balances lightweight performance, high classification accuracy and interpretability. Its excellent polyp recognition performance reduces the risks of misdiagnosis and missed diagnosis in clinical screening, while the lightweight design and GradCAM-based interpretability break the key barriers to the clinical adoption of AI in gastric cancer early diagnosis, providing a feasible AI-assisted tool for gastric precancerous lesion screening.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1812093","type":"journal-article","created":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T11:03:28Z","timestamp":1780571008000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Lightweight deep learning model for gastrointestinal precancerous lesion screening with attention enhancement"],"prefix":"10.3389","volume":"13","author":[{"given":"Shuai","family":"Chen","sequence":"first","affiliation":[{"name":"College of Life Sciences\/College of Artificial Intelligence, North China University of Science and Technology","place":["Tangshan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingyao","family":"Cai","sequence":"additional","affiliation":[{"name":"College of Life Sciences\/College of Artificial Intelligence, North China University of Science and Technology","place":["Tangshan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhixiang","family":"Wu","sequence":"additional","affiliation":[{"name":"College of Life Sciences\/College of Artificial Intelligence, North China University of Science and Technology","place":["Tangshan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiangyu","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Life Sciences\/College of Artificial Intelligence, North China University of Science and Technology","place":["Tangshan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qing","family":"Wang","sequence":"additional","affiliation":[{"name":"College of Life Sciences\/College of Artificial Intelligence, North China University of Science and Technology","place":["Tangshan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liming","family":"Zhou","sequence":"additional","affiliation":[{"name":"College of Life Sciences\/College of Artificial Intelligence, North China University of Science and Technology","place":["Tangshan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,6,4]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"10","DOI":"10.3322\/caac.21871","article-title":"Cancer statistics, 2025","volume":"75","author":"Siegel","year":"2025","journal-title":"CA Cancer J Clin"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"26","DOI":"10.5114\/pg.2018.80001","article-title":"Epidemiology of gastric cancer: global trends, risk factors and prevention","volume":"14","author":"Rawla","year":"2018","journal-title":"Gastroenterol Rev"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"653","DOI":"10.1007\/s10120-018-0793-2","article-title":"Application of artificial intelligence using a convolutional neural network for detecting gastric cancer in endoscopic images","volume":"21","author":"Hirasawa","year":"2018","journal-title":"Gastric Cancer"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"5408","DOI":"10.3748\/wjg.v26.i36.5408","article-title":"Artificial intelligence in gastric cancer: application and future perspectives","volume":"26","author":"Ni","year":"2020","journal-title":"World J Gastroenterol"},{"key":"ref5","article-title":"Deep residual learning for image recognition","author":"He","year":"2015"},{"key":"ref6","first-page":"2261","author":"Huang","year":"2017"},{"key":"ref7","article-title":"Efficientnet: rethinking model scaling for convolutional neural networks","author":"Tan","year":"2019"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"1431912","DOI":"10.3389\/fonc.2024.1431912","article-title":"Advanced CNN models in gastric cancer diagnosis: enhancing endoscopic image analysis with deep transfer learning","volume":"14","author":"Bhardwaj","year":"2024","journal-title":"Front Oncol"},{"key":"ref9","first-page":"4510","author":"Sandler","year":"2018"},{"key":"ref10","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1007\/978-3-030-01264-9_8","volume-title":"Computer Vision \u2013 ECCV 2018","author":"Ma","year":"2018"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"2718","DOI":"10.3390\/diagnostics12112718","article-title":"Gestronet: a framework of saliency estimation and optimal deep learning features based gastrointestinal diseases detection and classification","volume":"12","author":"Attique","year":"2022","journal-title":"Diagnostics"},{"key":"ref12","first-page":"164","author":"Pogorelov","year":"2017"},{"key":"ref13","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1007\/978-3-030-01234-2_1","volume-title":"Computer Vision \u2013 ECCV 2018","author":"Woo","year":"2018"},{"key":"ref14","first-page":"11531","author":"Wang","year":"2020"},{"key":"ref15","first-page":"7354","author":"Zhang","year":"2019"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"379","DOI":"10.3390\/sym15020379","article-title":"A deep-learning approach for identifying and classifying digestive diseases","volume":"15","author":"Vinoth","year":"2023","journal-title":"Symmetry"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1007\/s11263-019-01228-7","article-title":"Grad-cam: visual explanations from deep networks via gradient-based localization","volume":"128","author":"Selvaraju","year":"2020","journal-title":"Int J Comput Vis"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"482","DOI":"10.1111\/jgh.15190","article-title":"Application of artificial intelligence using a convolutional neural network for diagnosis of early gastric cancer based on magnifying endoscopy with narrow-band imaging","volume":"36","author":"Ueyama","year":"2020","journal-title":"J Gastroenterol Hepatol"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"108827","DOI":"10.1016\/j.patcog.2022.108827","article-title":"Gashis-transformer: a multi-scale visual transformer approach for gastric histopathological image detection","volume":"130","author":"Chen","year":"2022","journal-title":"Pattern Recogn"},{"key":"ref20","first-page":"207","author":"Song","year":"2018"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"4294","DOI":"10.1038\/s41467-020-18147-8","article-title":"Clinically applicable histopathological diagnosis system for gastric cancer detection using deep learning","volume":"11","author":"Reda","year":"2020","journal-title":"Nat Commun"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"51761","DOI":"10.1109\/ACCESS.2020.2979599","article-title":"Early prediction of the 2019 novel coronavirus outbreak in China's mainland based on a simple mathematical model","volume":"8","author":"Zhao","year":"2020","journal-title":"IEEE Access"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"11153","DOI":"10.1007\/s00500-022-07246-x","article-title":"Deep transfer learning techniques with hybrid optimization in early prediction and diagnosis of different types of oral cancer","volume":"26","author":"Bhardwaj","year":"2022","journal-title":"Soft Comput"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"7847","DOI":"10.1038\/s41598-024-58361-8","article-title":"Early gastric cancer detection and lesion segmentation based on deep learning and gastroscopic images","volume":"14","author":"Zhang","year":"2024","journal-title":"Sci Rep"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"2718","DOI":"10.3390\/diagnostics12112718","article-title":"GestroNet: a framework of saliency estimation and optimal deep learning features based gastrointestinal diseases detection and classification","volume":"12","author":"Khan","year":"2022","journal-title":"Diagnostics"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1812093\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T11:03:29Z","timestamp":1780571009000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1812093\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,4]]},"references-count":25,"alternative-id":["10.3389\/fmed.2026.1812093"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1812093","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,4]]},"article-number":"1812093"}}