{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T16:52:09Z","timestamp":1783529529622,"version":"3.55.0"},"reference-count":34,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,3,24]],"date-time":"2025-03-24T00:00:00Z","timestamp":1742774400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,3,24]],"date-time":"2025-03-24T00:00:00Z","timestamp":1742774400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Inform Decis Mak"],"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:sec>\n            <jats:title>Objective<\/jats:title>\n            <jats:p>In the functional assessment of the esophagogastric junction (EGJ), the endoscopic Hill classification plays a pivotal role in classifying the morphology of the gastroesophageal flap valve (GEFV). This study aims to develop an artificial intelligence model for Hill classification to assist endoscopists in diagnosis, covering the entire process from model development, testing, interpretability analysis, to multi-terminal deployment.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Method<\/jats:title>\n            <jats:p>The study collected four datasets, comprising a total of 1143 GEFV images and 17 gastroscopic videos, covering Hill grades I, II, III, and IV. The images were preprocessed and enhanced, followed by transfer learning using a pretrained model based on CNN and Transformer architectures. The model training utilized a cross-entropy loss function, combined with the Adam optimizer, and implemented a learning rate scheduling strategy. When assessing model performance, metrics such as accuracy, precision, recall, and F1 score were considered, and the diagnostic accuracy of the AI model was compared with that of endoscopists using McNemar\u2019s test, with a <jats:italic>p<\/jats:italic>-value\u2009&lt;\u20090.05 indicating statistical significance. To enhance model transparency, various interpretability analysis techniques were used, including t-SNE, Grad-CAM, and SHAP. Finally, the model was converted into ONNX format and deployed on multiple device terminals.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Results<\/jats:title>\n            <jats:p>Compared through performance metrics, the EfficientNet-Hill model surpassed other CNN and Transformer models, achieving an accuracy of 83.32% on the external test set, slightly lower than senior endoscopists (86.51%) but higher than junior endoscopists (75.82%). McNemar\u2019s test showed a significant difference in classification performance between the model and junior endoscopists (<jats:italic>p<\/jats:italic>\u2009&lt;\u20090.05), but no significant difference between the model and senior endoscopists (<jats:italic>p<\/jats:italic>\u2009\u2265\u20090.05). Additionally, the model reached precision, recall, and F1 scores of 84.81%, 83.32%, and 83.95%, respectively. Despite its overall excellent performance, there were still misclassifications. Through interpretability analysis, key areas of model decision-making and reasons for misclassification were identified. Finally, the model achieved real-time automatic Hill classification at over 50fps on multiple platforms.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Conclusion<\/jats:title>\n            <jats:p>By employing deep learning to construct the EfficientNet-Hill AI model, automated Hill classification of GEFV morphology was achieved, aiding endoscopists in improving diagnostic efficiency and accuracy in endoscopic grading, and facilitating the integration of Hill classification into routine endoscopic reports and GERD assessments.<\/jats:p>\n          <\/jats:sec>","DOI":"10.1186\/s12911-025-02973-1","type":"journal-article","created":{"date-parts":[[2025,3,24]],"date-time":"2025-03-24T22:40:42Z","timestamp":1742856042000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Constructing an artificial intelligence-assisted system for the assessment of gastroesophageal valve function based on the hill classification (with video)"],"prefix":"10.1186","volume":"25","author":[{"given":"Jian","family":"Chen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ganhong","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kaijian","family":"Xia","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhenni","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luojie","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaodan","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,3,24]]},"reference":[{"issue":"1","key":"2973_CR1","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1007\/s12664-020-01135-7","volume":"40","author":"UC Ghoshal","year":"2021","unstructured":"Ghoshal UC, Singh R, Rai S. Prevalence and risk factors of gastroesophageal reflux disease in a rural Indian population. Indian J Gastroenterol Off J Indian Soc Gastroenterol. 2021;40(1):56\u201364.","journal-title":"Indian J Gastroenterol Off J Indian Soc Gastroenterol."},{"issue":"2","key":"2973_CR2","first-page":"59","volume":"11","author":"AM Altwigry","year":"2017","unstructured":"Altwigry AM, Almutairi MS, Ahmed M. Gastroesophageal reflux disease prevalence among school teachers of Saudi Arabia and its impact on their daily life activities. Int J Health Sci. 2017;11(2):59\u201364.","journal-title":"Int J Health Sci."},{"key":"2973_CR3","doi-asserted-by":"crossref","first-page":"1351","DOI":"10.1136\/gutjnl-2017-314722","volume":"67","author":"CP Gyawali","year":"2018","unstructured":"Gyawali CP, Kahrilas PJ, Savarino E, Zerbib F, Mion F, Smout AJPM, Vaezi M, Sifrim D, Fox MR, Vela MF, et al. Modern diagnosis of GERD: the Lyon Consensus. Gut. 2018;67:1351\u201362.","journal-title":"Gut."},{"issue":"1","key":"2973_CR4","doi-asserted-by":"crossref","first-page":"118","DOI":"10.1186\/s12876-017-0693-7","volume":"17","author":"C Xie","year":"2017","unstructured":"Xie C, Li Y, Zhang N, Xiong L, Chen M, Xiao Y. Gastroesophageal flap valve reflected EGJ morphology and correlated to acid reflux. BMC Gastroenterol. 2017;17(1):118.","journal-title":"BMC Gastroenterol."},{"issue":"2","key":"2973_CR5","first-page":"519","volume":"64","author":"JA Tocornal","year":"1968","unstructured":"Tocornal JA, Snow HD, Fonkalsrud EW. A mucosol flap valve mechanism to prevent gastroesophageal reflux and esophagitis. Surgery. 1968;64(2):519\u201323.","journal-title":"Surgery."},{"issue":"1","key":"2973_CR6","first-page":"25","volume":"153","author":"KB Thor","year":"1987","unstructured":"Thor KB, Hill LD, Mercer DD, Kozarek RD. Reappraisal of the flap valve mechanism in the gastroesophageal junction. A study of a new valvuloplasty procedure in cadavers. Acta Chirurgica Scand. 1987;153(1):25\u201328.","journal-title":"Acta Chirurgica Scand."},{"issue":"5","key":"2973_CR7","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1016\/S0016-5107(96)70006-8","volume":"44","author":"LD Hill","year":"1996","unstructured":"Hill LD, Kozarek RA, Kraemer SJ, Aye RW, Mercer CD, Low DE, Pope CEN. The gastroesophageal flap valve: in vitro and in vivo observations. Gastrointestinal Endoscopy. 1996;44(5):541\u201347.","journal-title":"Gastrointestinal Endoscopy."},{"issue":"6","key":"2973_CR8","doi-asserted-by":"crossref","first-page":"533","DOI":"10.1007\/s00535-002-1100-9","volume":"38","author":"Y Fujiwara","year":"2003","unstructured":"Fujiwara Y, Higuchi K, Shiba M, Watanabe T, Tominaga K, Oshitani N, Matsumoto T, Arakawa T. Association between gastroesophageal flap valve, reflux esophagitis, Barrett\u2019s epithelium, and atrophic gastritis assessed by endoscopy in Japanese patients. J Gastroenterol. 2003;38(6):533\u201339.","journal-title":"J Gastroenterol."},{"issue":"1","key":"2973_CR9","first-page":"100","volume":"33","author":"Y Koya","year":"2021","unstructured":"Koya Y, Shibata M, Watanabe T, Kumei S, Miyagawa K, Oe S, Honma Y, Kume K, Yoshikawa I, Harada M. Influence of gastroesophageal flap valve on esophageal variceal bleeding in patients with liver cirrhosis. Digestive Endoscopy Off J Jpn Gastroenterological Endoscopy Soc. 2021;33(1):100\u201309.","journal-title":"Digestive Endoscopy Off J Jpn Gastroenterological Endoscopy Soc."},{"key":"2973_CR10","doi-asserted-by":"crossref","first-page":"15744","DOI":"10.1038\/s41598-019-52349-5","volume":"9","author":"W Wu","year":"2019","unstructured":"Wu W, Li L, Qu C, Wang M, Liang S, Gao X, Bao X, Wang L, Liu H, Han H, et al. Reflux finding score is associated with gastroesophageal flap valve status in patients with laryngopharyngeal reflux disease: a retrospective study. Sci Rep-UK. 2019;9:15744.","journal-title":"Sci Rep-UK."},{"issue":"2","key":"2973_CR11","doi-asserted-by":"crossref","first-page":"226","DOI":"10.5056\/jnm17088","volume":"24","author":"DT Quach","year":"2018","unstructured":"Quach DT, Nguyen TT, Hiyama T. Abnormal gastroesophageal flap valve is associated with high gastresophageal reflux disease questionnaire score and the severity of gastroesophageal reflux disease in Vietnamese patients with upper gastrointestinal symptoms. J Neurogastroenterology. 2018;24(2):226\u201332.","journal-title":"J Neurogastroenterology."},{"issue":"1","key":"2973_CR12","doi-asserted-by":"crossref","first-page":"151","DOI":"10.1007\/s10620-020-06146-0","volume":"66","author":"A Osman","year":"2021","unstructured":"Osman A, Albashir MM, Nandipati K, Walters RW, Chandra S. Esophagogastric junction morphology on hill\u2019s classification predicts gastroesophageal reflux with good accuracy and consistency. Digestive Dis Sci. 2021;66(1):151\u201359.","journal-title":"Digestive Dis Sci."},{"issue":"12","key":"2973_CR13","doi-asserted-by":"crossref","first-page":"660","DOI":"10.21037\/atm-22-2071","volume":"10","author":"X Sui","year":"2022","unstructured":"Sui X, Gao X, Zhang L, Zhang B, Zhong C, Chen Y, Wang X, Li D, Wu W, Li L. Clinical efficacy of endoscopic antireflux mucosectomy vs. Stretta radiofrequency in the treatment of gastroesophageal reflux disease: a retrospective, single-center cohort study. Ann Transl Med. 2022;10(12):660.","journal-title":"Ann Transl Med."},{"key":"2973_CR14","doi-asserted-by":"crossref","unstructured":"Wang C, Chiu Y, Chen W, Yang T, Tsai M, Tseng M. A deep learning model for classification of endoscopic gastroesophageal reflux disease. Int J Environ Res Public Health. 2021;18(5).","DOI":"10.3390\/ijerph18052428"},{"key":"2973_CR15","doi-asserted-by":"crossref","unstructured":"Yen H, Tsai H, Wang C, Tsai M, Tseng M. An improved endoscopic automatic classification model for gastroesophageal reflux disease using deep learning integrated machine learning. Diagnostics (Basel, Switzerland). 2022;12(11).","DOI":"10.3390\/diagnostics12112827"},{"issue":"3","key":"2973_CR16","doi-asserted-by":"crossref","first-page":"e282532","DOI":"10.1371\/journal.pone.0282532","volume":"18","author":"C Athalye","year":"2023","unstructured":"Athalye C, Arnaout R.Domain-guided data augmentation for deep learning on medical imaging. PLoS One. 2023;18(3):e282532.","journal-title":"PLoS One."},{"key":"2973_CR17","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. Las Vegas, NV, USA; 2016. p. 770\u201378.","DOI":"10.1109\/CVPR.2016.90"},{"issue":"4","key":"2973_CR18","doi-asserted-by":"crossref","first-page":"3609","DOI":"10.1007\/s12652-021-03488-z","volume":"14","author":"M Bansal","year":"2023","unstructured":"Bansal M, Kumar M, Sachdeva M, Mittal A.Transfer learning for image classification using VGG19: Caltech-101 image data set. J Ambient Intell Hum Comput. 2023;14(4):3609\u201320.","journal-title":"J Ambient Intell Hum Comput."},{"key":"2973_CR19","doi-asserted-by":"crossref","unstructured":"Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR): 2017\/1\/1. 2017. Vol. 2017, p. 2261\u201369.","DOI":"10.1109\/CVPR.2017.243"},{"key":"2973_CR20","doi-asserted-by":"crossref","unstructured":"Chen L, Zhu Y, Papandreou G, Schroff F, Adam H. Encoder-decoder with atrous separable convolution for semantic image segmentation. In: Computer Vision \u2013 ECCV 2018. 2018. p. 833\u201351.","DOI":"10.1007\/978-3-030-01234-2_49"},{"issue":"9","key":"2973_CR21","doi-asserted-by":"crossref","first-page":"10870","DOI":"10.1109\/TPAMI.2023.3268446","volume":"45","author":"T Yao","year":"2023","unstructured":"Yao T, Li Y, Pan Y, Wang Y, Zhang X, Mei T. Dual vision transformer. IEEE Trans Pattern Anal Mach Intell. 2023;45(9):10870\u201382.","journal-title":"IEEE Trans Pattern Anal Mach Intell."},{"key":"2973_CR22","doi-asserted-by":"crossref","unstructured":"Liu Z, Lin Y, Cao Y, Hu H, Wei Y, Zhang Z, Lin S, Guo B. Swin transformer: hierarchical vision transformer using shifted windows. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV): 2021\/1\/1. 2021. Vol. 2021, p. 9992\u201310002.","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"2973_CR23","doi-asserted-by":"crossref","unstructured":"Wu H, Xiao B, Codella N, Liu M, Dai X, Yuan L, Zhang L. CvT: introducing convolutions to vision transformers. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV): 2021\/1\/1. 2021. Vol. 2021, p. 22\u201331.","DOI":"10.1109\/ICCV48922.2021.00009"},{"key":"2973_CR24","doi-asserted-by":"crossref","first-page":"109098","DOI":"10.1016\/j.jneumeth.2021.109098","volume":"353","author":"Y Zhang","year":"2021","unstructured":"Zhang Y, Hong D, McClement D, Oladosu O, Pridham G, Slaney G. Grad-CAM helps interpret the deep learning models trained to classify multiple sclerosis types using clinical brain magnetic resonance imaging. J Neurosci Methods. 2021;353:109098.","journal-title":"J Neurosci Methods."},{"issue":"15","key":"2973_CR25","doi-asserted-by":"crossref","first-page":"154108","DOI":"10.1063\/5.0087310","volume":"156","author":"T Kikutsuji","year":"2022","unstructured":"Kikutsuji T, Mori Y, Okazaki K, Mori T, Kim K, Matubayasi N. Explaining reaction coordinates of alanine dipeptide isomerization obtained from deep neural networks using Explainable Artificial Intelligence (XAI). J Chem Phys. 2022;156(15):154108.","journal-title":"J Chem Phys."},{"issue":"2","key":"2973_CR26","doi-asserted-by":"crossref","first-page":"313","DOI":"10.1137\/18M1216134","volume":"1","author":"GC Linderman","year":"2019","unstructured":"Linderman GC, Steinerberger S. Clustering with t-SNE, provably. SIAM J Math Data Sci. 2019;1(2):313\u201332.","journal-title":"SIAM J Math Data Sci."},{"key":"2973_CR27","doi-asserted-by":"crossref","unstructured":"Li P, Wang X, Huang K, Huang Y, Li S, Iqbal M. Multi-model running latency optimization in an edge computing paradigm. Sensors (Basel, Switzerland). 2022;22(16).","DOI":"10.3390\/s22166097"},{"issue":"3","key":"2973_CR28","doi-asserted-by":"crossref","first-page":"E311","DOI":"10.1055\/s-0042-101021","volume":"4","author":"I Hansdotter","year":"2016","unstructured":"Hansdotter I, Bj\u00f6r O, Andreasson A, Agreus L, Hellstr\u00f6m P, Forsberg A, Talley NJ, Vieth M, Wallner B. Hill classification is superior to the axial length of a hiatal hernia for assessment of the mechanical anti-reflux barrier at the gastroesophageal junction. Endoscopy Int Open. 2016;4(3):E311\u2013E317.","journal-title":"Endoscopy Int Open."},{"issue":"2","key":"2973_CR29","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1111\/nmo.12507","volume":"27","author":"PW Weijenborg","year":"2015","unstructured":"Weijenborg PW, van Hoeij FB, Smout AJPM, Bredenoord AJ. Accuracy of hiatal hernia detection with esophageal high-resolution manometry. Neurogastroenterology Motil. 2015;27(2):293\u201399.","journal-title":"Neurogastroenterology Motil."},{"key":"2973_CR30","doi-asserted-by":"crossref","first-page":"102802","DOI":"10.1016\/j.media.2023.102802","volume":"88","author":"F Shamshad","year":"2023","unstructured":"Shamshad F, Khan S, Zamir SW, Khan MH, Hayat M, Khan FS, Fu H. Transformers in medical imaging: a survey. Med Image Anal. 2023;88:102802.","journal-title":"Med Image Anal."},{"key":"2973_CR31","doi-asserted-by":"crossref","first-page":"102444","DOI":"10.1016\/j.media.2022.102444","volume":"79","author":"X Chen","year":"2022","unstructured":"Chen X, Wang X, Zhang K, Fung K, Thai TC, Moore K, Mannel RS, Liu H, Zheng B, Qiu Y. Recent advances and clinical applications of deep learning in medical image. Med Image Anal. 2022;79:102444.","journal-title":"Med Image Anal."},{"key":"2973_CR32","doi-asserted-by":"crossref","first-page":"3514807","DOI":"10.1155\/2022\/3514807","volume":"2022","author":"A Booysens","year":"2022","unstructured":"Booysens A, Viriri S. Exploration of ear biometrics using EfficientNet. Comput Intell Neurosci. 2022;2022:3514807.","journal-title":"Comput Intell Neurosci."},{"key":"2973_CR33","doi-asserted-by":"crossref","unstructured":"Rahhal MMA, Bazi Y, Jomaa RM, Zuair M, Melgani F. Contrasting EfficientNet, ViT, and gMLP for COVID-19 detection in ultrasound. J Pers Med. 2022;12(10).","DOI":"10.3390\/jpm12101707"},{"issue":"1","key":"2973_CR34","doi-asserted-by":"crossref","first-page":"250","DOI":"10.1109\/TPAMI.2023.3327511","volume":"46","author":"Y Yuan","year":"2024","unstructured":"Yuan Y, Liang W, Ding H, Liang Z, Zhang C, Hu H.Expediting large-scale vision transformer for dense prediction without. IEEE Trans Pattern Anal Mach Intell. 2024;46(1):250\u201366.","journal-title":"IEEE Trans Pattern Anal Mach Intell."}],"container-title":["BMC Medical Informatics and Decision Making"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-025-02973-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12911-025-02973-1\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-025-02973-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,24]],"date-time":"2025-03-24T22:42:01Z","timestamp":1742856121000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcmedinformdecismak.biomedcentral.com\/articles\/10.1186\/s12911-025-02973-1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,24]]},"references-count":34,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["2973"],"URL":"https:\/\/doi.org\/10.1186\/s12911-025-02973-1","relation":{},"ISSN":["1472-6947"],"issn-type":[{"value":"1472-6947","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,24]]},"assertion":[{"value":"4 January 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 March 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"24 March 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study was approved by the Ethics Committee of The Changshu Hospital Affiliated to Soochow University (IRB approval number L2023047). Due to the study\u2019s non-interventional retrospective design, written informed patient consent was waived by the IRB.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable. This retrospective study did not include any personal identifiable information, and the requirement for written informed consent for publication was waived by the Ethics Committee of The Changshu Hospital Affiliated to Soochow University (IRB approval number L2023047).","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"144"}}