{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T06:41:05Z","timestamp":1776235265793,"version":"3.50.1"},"reference-count":40,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T00:00:00Z","timestamp":1776211200000},"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:p>Gastrointestinal cancer (GC) is particularly malignant as they tend to progress slowly before advanced stages due to the forecast of early-specific symptoms. The heterogeneous properties of GCs require extremely precise and sensitive diagnostic techniques that can integrate in-depth structural information and surface-level data to distinguish cancer severity grades, thereby significantly lowering mortality rates. Deep learning (DL) algorithms are crucial in the classification of these various GC grades. However, the algorithms cannot detect interpretability and are characterized by a high probability of false alarm rates in detecting the underlying acute relationship between the medical images. Additionally, existing systems lack language-level transparency, preventing them from generating user-based narrative diagnostic explanations consistent with medical standards. To address this aforementioned challenge, this research study introduces a novel explainable LLM (X-LLM) based DL framework, which overcomes the drawbacks of existing DL algorithms. The suggested framework uses the ensemble transformer architectures that combine the clinical features by integrating the endoscopy and computer tomography (CT) scan images for enhancing the performance in detecting the different severity grades of GCs. The proposed system uses several components: (1) heterogeneous image collection; (2) image pre-processing; (3) ensemble networks; (4) interpretability analysis; and (5) user-interaction module. The extensive experiments are conducted utilizing two different datasets, such as Kvasir and TCIA CT (TCGA-STAD) scan images. The severity annotation of both datasets was carried out by experienced medical doctors, including endoscopists. Several evaluation metrics, including accuracy, precision, and recall, are measured and benchmarked against other learning networks. The experimental findings demonstrate the enhanced performance of the proposed framework over the existing models by achieving the accuracy, precision, recall, and F1-score values of 0.99, 0.997, 0.99, and 0.99, respectively. Furthermore, different LLM models such as GPT4.0, GPT3.5, LLAMA, and GEMINI are integrated, and their mode of interaction is also analyzed with SHAP measurements. The suggested framework demonstrates its strong potential by enhancing diagnostic performance, achieving high performance with user-interacted clinical treatment outcomes.<\/jats:p>","DOI":"10.3389\/fmed.2026.1789769","type":"journal-article","created":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T05:41:05Z","timestamp":1776231665000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Building novel LLM-enabled explainable ensemble transformer models combining endoscopic and CT images for discriminating the different grades of gastrointestinal cancers"],"prefix":"10.3389","volume":"13","author":[{"given":"Ali","family":"Algarni","sequence":"first","affiliation":[{"name":"Department of Informatics and Computer Systems, College of Computer Science, King Khalid University","place":["Abha, Asir, Saudi Arabia"]},{"name":"Center for Artificial Intelligence, King Khalid University","place":["Abha, Asir, Saudi Arabia"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ahood A.","family":"Al-Eidan","sequence":"additional","affiliation":[{"name":"Department of Biology, College of Science, Imam Abdulrahman Bin Faisal University","place":["Dammam, Saudi Arabia"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Surbhi B.","family":"Khan","sequence":"additional","affiliation":[{"name":"School of Science, Engineering and Environment, University of Salford","place":["Manchester, United Kingdom"]},{"name":"Centre for Research Impact & Outcome, Chitkara University","place":["Rajpura, Punjab, India"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Marwah A.","family":"Halwani","sequence":"additional","affiliation":[{"name":"Department of Management Information Systems, College of Business, King Abdulaziz University","place":["Rabigh, Saudi Arabia"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ahlam","family":"Almusharraf","sequence":"additional","affiliation":[{"name":"Department of Management, College of Business Administration, Princess Nourah Bint Abdulrahman University","place":["Riyadh, Saudi Arabia"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yongwon","family":"Cho","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Soonchunhyang University","place":["Asan, Republic of Korea"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yunyoung","family":"Nam","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Soonchunhyang University","place":["Asan, Republic of Korea"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,4,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"335","DOI":"10.1053\/j.gastro.2020.02.068","article-title":"Global burden of 5 major types of gastrointestinal cancer","volume":"159","author":"Arnold","year":"2020","journal-title":"Gastroenterology"},{"key":"B2","volume-title":"Global Cancer Observatory: Cancer Today","author":"Ferlay","year":"2020"},{"key":"B3","doi-asserted-by":"publisher","first-page":"744","DOI":"10.1053\/j.gastro.2020.10.007","article-title":"Worldwide burden of, risk factors for, and trends in pancreatic cancer","volume":"160","author":"Huang","year":"2021","journal-title":"Gastroenterology."},{"key":"B4","doi-asserted-by":"publisher","first-page":"404","DOI":"10.1002\/cac2.12143","article-title":"Predictions of mortality related to four major cancers in China, 2020 to 2030","volume":"41","author":"Li","year":"2021","journal-title":"Cancer Commun."},{"key":"B5","doi-asserted-by":"publisher","first-page":"105","DOI":"10.3390\/biom10010105","article-title":"Phytochemicals and gastrointestinal cancer: cellular mechanisms and effects to change cancer progression","volume":"10","author":"Al-Ishaq","year":"2020","journal-title":"Biomolecules."},{"key":"B6","doi-asserted-by":"publisher","first-page":"245023","DOI":"10.1088\/1361-6560\/ad0a5a","article-title":"Deep learning for fast super-resolution ultrasound microvessel imaging","volume":"68","author":"Luan","year":"2023","journal-title":"Phys Med Biol."},{"key":"B7","doi-asserted-by":"publisher","first-page":"205002","DOI":"10.1088\/1361-6560\/acf98f","article-title":"Deep learning for fast denoising filtering in ultrasound localization microscopy","volume":"68","author":"Yu","year":"2023","journal-title":"Phys Med Biol."},{"key":"B8","doi-asserted-by":"publisher","first-page":"10965","DOI":"10.1109\/TMM.2024.3428349","article-title":"CenterFormer: a novel cluster center enhanced transformer for unconstrained dental plaque segmentation","volume":"26","author":"Song","year":"2024","journal-title":"IEEE Trans Multimed."},{"key":"B9","doi-asserted-by":"publisher","first-page":"230212","DOI":"10.29026\/oea.2024.230212","article-title":"Large-field objective lens for multi-wavelength microscopy at mesoscale and submicron resolution","volume":"7","author":"Xu","year":"2024","journal-title":"Opto-Electron Adv."},{"key":"B10","doi-asserted-by":"publisher","first-page":"119126","DOI":"10.1016\/j.jep.2024.119126","article-title":"Chrysotoxine regulates ferroptosis and the PI3K\/AKT\/mTOR pathway to prevent cervical cancer","volume":"338","author":"Zhou","year":"2025","journal-title":"J Ethnopharmacol."},{"key":"B11","doi-asserted-by":"publisher","first-page":"24","DOI":"10.36922\/arnm.8350","article-title":"Current status and future potential of radiomics in the management of patients with gastric cancer","volume":"3","author":"Zheng","year":"2025","journal-title":"Adv Radiother Nucl Med."},{"key":"B12","doi-asserted-by":"publisher","first-page":"876","DOI":"10.1038\/s42256-024-00868-w","article-title":"A transformer-based weakly supervised computational pathology method for clinical-grade diagnosis and molecular marker discovery of gliomas","volume":"6","author":"Jiang","year":"2024","journal-title":"Nat Mach Intell."},{"key":"B13","doi-asserted-by":"publisher","first-page":"164","DOI":"10.2174\/1574893616666210708143556","article-title":"Cervical cancer metastasis and recurrence risk prediction based on deep convolutional neural network","volume":"17","author":"Ye","year":"2022","journal-title":"Curr Bioinform."},{"key":"B14","doi-asserted-by":"publisher","first-page":"1652","DOI":"10.7150\/ijbs.92988","article-title":"The burgeoning significance of liquid-liquid phase separation in the pathogenesis and therapeutics of cancers","volume":"20","author":"Xu","year":"2024","journal-title":"Int J Biol Sci"},{"key":"B15","first-page":"1","article-title":"An ensemble-based model of detecting plant disease using CNN and random forest","volume-title":"2023 6th International Conference on Information Systems and Computer Networks (ISCON)","author":"Saxena","year":"2023"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1596085","DOI":"10.3389\/fimmu.2025.1596085","article-title":"Interpretable deep learning for gastric cancer detection: a fusion of AI architectures and explainability analysis","volume":"16","author":"Ma","year":"2025","journal-title":"Front Immunol."},{"key":"B17","doi-asserted-by":"publisher","first-page":"362","DOI":"10.1186\/s12967-025-06363-z","article-title":"Multi-cohort study in gastric cancer to develop CT-based radiomic models to predict pathological response to neoadjuvant immunotherapy","volume":"23","author":"Huang","year":"2025","journal-title":"J Transl Med."},{"key":"B18","doi-asserted-by":"publisher","first-page":"e75556","DOI":"10.2196\/75556","article-title":"Automated esophageal cancer staging from free-text radiology reports: large language model evaluation study","volume":"13","author":"Yao","year":"2025","journal-title":"JMIR Med Inform"},{"key":"B19","doi-asserted-by":"publisher","first-page":"109948","DOI":"10.3748\/wjg.v31.i31.109948","article-title":"DeepGut: a collaborative multimodal large language model framework for digestive disease assisted diagnosis and treatment","volume":"31","author":"Wan","year":"2025","journal-title":"World J Gastroenterol"},{"key":"B20","doi-asserted-by":"publisher","first-page":"21777","DOI":"10.1007\/s11042-024-19816-2","article-title":"Histopathology image analysis for gastric cancer detection: a hybrid deep learning and catboost approach","volume":"84","author":"Khayatian","year":"2025","journal-title":"Multimed Tools Appl"},{"key":"B21","doi-asserted-by":"publisher","first-page":"13087","DOI":"10.1038\/s41598-025-97256-0","article-title":"An interpretable framework for gastric cancer classification using multi-channel attention mechanisms and transfer learning approach on histopathology images","volume":"15","author":"Zubair","year":"2025","journal-title":"Sci Rep."},{"key":"B22","doi-asserted-by":"publisher","first-page":"104885","DOI":"10.1016\/j.jbi.2025.104885","article-title":"GutGPT: a multidimensional knowledge-enhanced large language model for gastrointestinal medicine","volume":"169","author":"Zhang","year":"2025","journal-title":"J Biomed Inform"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1583514","DOI":"10.3389\/fmed.2025.1583514","article-title":"Enhancing gastroenterology with multimodal learning: The role of large language model chatbots in digestive endoscopy","volume":"12","author":"Qin","year":"2025","journal-title":"Front Med"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1349373","DOI":"10.3389\/fmed.2024.1349373","article-title":"Explainable AI-driven model for gastrointestinal cancer classification","volume":"11","author":"Binzagr","year":"2024","journal-title":"Front Med"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1746","DOI":"10.3390\/diagnostics14161746","article-title":"Gastric cancer detection with ensemble learning on digital pathology: use case of gastric cancer on GasHisSDB dataset","volume":"14","author":"Mudavadkar","year":"2024","journal-title":"Diagnostics"},{"key":"B26","unstructured":""},{"key":"B27","doi-asserted-by":"crossref","first-page":"8107","DOI":"10.1109\/CVPR42600.2020.00813","article-title":"Analyzing and improving the image quality of StyleGAN","volume-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Karras","year":"2020"},{"key":"B28","unstructured":"Empirical evaluation of gated recurrent neural networks on sequence modeling\n          \n          \n            \n              Chung\n              J\n            \n            \n              Gulcehre\n              C\n            \n            \n              Cho\n              K\n            \n            \n              Bengio\n              Y\n            \n          \n          arXiv\n          \n          2014"},{"key":"B29","doi-asserted-by":"publisher","first-page":"9031","DOI":"10.3390\/app13159031","article-title":"Localization and classification of gastrointestinal tract disorders using explainable AI from endoscopic images","volume":"13","author":"Nouman Noor","year":"2023","journal-title":"Appl Sci."},{"key":"B30","doi-asserted-by":"publisher","first-page":"41611","DOI":"10.1007\/s11042-021-11007-7","article-title":"A review on extreme learning machine","volume":"81","author":"Wang","year":"2022","journal-title":"Multimed Tools Appl."},{"key":"B31","doi-asserted-by":"publisher","first-page":"7710","DOI":"10.3390\/s24237710","article-title":"Automated detection of gastrointestinal diseases using Resnet50*-based explainable deep feature engineering model with endoscopy images","volume":"24","author":"Cambay","year":"2024","journal-title":"Sensors"},{"key":"B32","doi-asserted-by":"publisher","first-page":"106924","DOI":"10.1016\/j.cmpb.2022.106924","article-title":"StoHisNet: a hybrid multi-classification model with CNN and transformer for gastric pathology images","volume":"221","author":"Fu","year":"2022","journal-title":"Comput Methods Programs Biomed."},{"key":"B33","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/EUVIP61797.2024.10772988","article-title":"Advancing gastrointestinal disease diagnosis: a fine-grained approach using swin transformer and explainable AI techniques","volume-title":"2024 12th European Workshop on Visual Information Processing (EUVIP)","author":"Fahad","year":"2024"},{"key":"B34","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ICCP63557.2024.10793024","article-title":"Explainable cancer segmentation through classification","author":"Dug\u0103e\u015fescu","year":"2024","journal-title":"2024 IEEE 20th International Conference on Intelligent Computer Communication and Processing (ICCP)"},{"key":"B35","doi-asserted-by":"crossref","first-page":"619","DOI":"10.1109\/ICMLCA63499.2024.10753879","article-title":"Leveraging large language models: enhancing retrieval-augmented generation with ScaNN and gemma for superior AI response","volume-title":"2024 5th International Conference on Machine Learning and Computer Application (ICMLCA)","author":"Gao","year":"2024"},{"key":"B36","doi-asserted-by":"crossref","first-page":"1081","DOI":"10.1109\/CAI59869.2024.00108","article-title":"Performance analysis of Llama 2 among other LLMs","volume-title":"2024 IEEE Conference on Artificial Intelligence (CAI)","author":"Huang","year":"2024"},{"key":"B37","article-title":"Capabilities of gemini models in medicine","author":"Saab","year":"2024","journal-title":"arXiv"},{"key":"B38","doi-asserted-by":"publisher","first-page":"395","DOI":"10.1109\/CSR61664.2024.10679444","article-title":"Can a llama be a watchdog?","author":"Curto","year":"2024"},{"key":"B39","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/IISEC59749.2023.10391005","article-title":"Generative pre-trained transformer (GPT) models for irony detection and classification","volume-title":"2023 4th International Informatics and Software Engineering Conference (IISEC)","author":"Aytekin","year":"2023"},{"key":"B40","first-page":"383","article-title":"Large language model (LLM) & GPT, a monolithic study in generative AI","volume-title":"2023 Congress in Computer Science, Computer Engineering, & Applied Computing (CSCE)","author":"Mohammad","year":"2023"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1789769\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T05:41:09Z","timestamp":1776231669000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1789769\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,15]]},"references-count":40,"alternative-id":["10.3389\/fmed.2026.1789769"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1789769","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,15]]},"article-number":"1789769"}}