{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T17:51:50Z","timestamp":1785433910485,"version":"3.56.0"},"reference-count":32,"publisher":"MDPI AG","issue":"12","license":[{"start":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T00:00:00Z","timestamp":1781222400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100005058","name":"NED University of Engineering and Technology","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100005058","id-type":"DOI","asserted-by":"crossref"}]},{"id":[{"id":"https:\/\/ror.org\/05db8zr24","id-type":"ROR","asserted-by":"publisher"}]},{"name":"Sindh Institute of Advanced Endoscopy and Gastroenterology"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis (PEP) is a clinically significant complication of ERCP, occurring in approximately 2\u201310% of general cases and at higher rates in high-risk patients. Early prediction of PEP risk may support timely intervention and improved patient management. This retrospective single-center study comparatively evaluated tabular clinical data, endoscopic image data, and multimodal fusion approaches for PEP prediction. Methods: Retrospective data collected from the Sindh Institute of Advanced Endoscopy and Gastroenterology were analyzed using machine learning and deep learning techniques. XGBoost(version 3.2.0) was applied to tabular clinical data, while EfficientNet-B0, ResNet50, and DenseNet201 were used for endoscopic image analysis. A multimodal contrastive learning (MMCL)-based framework combining ResNet50 image features with multilayer perceptron (MLP)-based tabular features was additionally implemented for binary PEP prediction. Class imbalance mitigation techniques, including data augmentation and balancing strategies, were applied during training. Model performance was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, F1-score, and precision. SHAP analysis was performed to identify important predictive features. Results: The tabular XGBoost model achieved the best predictive performance with an AUC of 0.95 and a sensitivity of 0.50, while five-fold cross-validation yielded an AUC of 0.79 and a sensitivity of 0.48. Among image-based models, ResNet50 achieved the highest performance, with an AUC of 0.76 and a sensitivity of 0.40. The multimodal model achieved an AUC of 0.57 and a sensitivity of 0.20. SHAP analysis identified cannulation time, ampulla type, and age as prominent features associated with PEP prediction. Conclusions: This exploratory study suggests that structured clinical data currently provide stronger predictive signals for PEP prediction than the available image and multimodal data within this limited cohort. The relatively low occurrence of PEP contributed to class imbalance despite mitigation strategies. Future multicenter studies with larger datasets, improved image availability, synthetic data generation, and advanced multimodal fusion techniques may improve predictive performance and clinical applicability.<\/jats:p>","DOI":"10.3390\/diagnostics16121824","type":"journal-article","created":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T00:44:58Z","timestamp":1781484298000},"page":"1824","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["AI-Based Prediction of Post-ERCP Pancreatitis: A Comparative Study Using Tabular, Image, and Multimodal Data"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-2318-6710","authenticated-orcid":false,"given":"Anum","family":"Jamil","sequence":"first","affiliation":[{"name":"Department of Computer Science and Information Technology, NED University of Engineering and Technology, Main University Road, Karachi 75270, Sindh, Pakistan"},{"name":"Sindh Institute of Advance Endoscopy and Gastroenterology, Heritage Building 1st Floor, Dr Ruth K.M. Pfau Civil Hospital, Baba-e-Urdu Road, Karachi 74400, Sindh, Pakistan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7386-6196","authenticated-orcid":false,"given":"Waseemullah","family":"Nazir","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Information Technology, NED University of Engineering and Technology, Main University Road, Karachi 75270, Sindh, Pakistan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0150-074X","authenticated-orcid":false,"given":"Abeer","family":"Altaf","sequence":"additional","affiliation":[{"name":"Sindh Institute of Advance Endoscopy and Gastroenterology, Heritage Building 1st Floor, Dr Ruth K.M. Pfau Civil Hospital, Baba-e-Urdu Road, Karachi 74400, Sindh, Pakistan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Saad Khalid","family":"Niaz","sequence":"additional","affiliation":[{"name":"Sindh Institute of Advance Endoscopy and Gastroenterology, Heritage Building 1st Floor, Dr Ruth K.M. Pfau Civil Hospital, Baba-e-Urdu Road, Karachi 74400, Sindh, Pakistan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,12]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"De, T., Du, G., Yin, H., Wang, H., Wang, W., Ma, T., Ma, J., Wang, H., and Wang, Q. (2025). Development and validation of a practical prediction model for post-ERCP pancreatitis using machine learning. Front. Surg., 12.","DOI":"10.3389\/fsurg.2025.1628956"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"111003","DOI":"10.4240\/wjgs.v17.i12.111003","article-title":"Prediction model for the occurrence of acute pancreatitis after endoscopic retrograde cholangiopancrea-tography based on multidimensional indicators","volume":"17","author":"Cao","year":"2025","journal-title":"World J. Gastrointest. Surg."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"657","DOI":"10.1055\/s-0042-108641","article-title":"Papillary cannulation and sphincter-otomy techniques at ERCP: European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline","volume":"48","author":"Testoni","year":"2016","journal-title":"Endoscopy"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1016\/j.dld.2022.10.005","article-title":"Machine learning for the prediction of post-ERCP pancreatitis risk: A proof-of-concept study","volume":"55","author":"Archibugi","year":"2023","journal-title":"Dig. Liver Dis."},{"key":"ref_5","first-page":"463","article-title":"Artificial intelligence in a prediction model for post-ERCP pancreatitis","volume":"36","author":"Takahashi","year":"2024","journal-title":"Dig. Endosc. Off. J. Jpn. Gastroenterol. Endosc. Soc."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"691","DOI":"10.1016\/j.gie.2024.03.031","article-title":"Duodenal papilla radiomics-based prediction model for post-ERCP pancreatitis using machine learning: A retrospective multicohort study","volume":"100","author":"Chen","year":"2024","journal-title":"Gastrointest. Endosc."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"8381","DOI":"10.1038\/s41598-021-87737-3","article-title":"Artificial intelligence-assisted analysis of endoscopic retrograde cholangiopancreatography image for identifying ampulla and difficulty of selective cannulation","volume":"11","author":"Kim","year":"2021","journal-title":"Sci. Rep."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"504","DOI":"10.1177\/2050640616674837","article-title":"Scandinavian Association for Digestive Endos-copy (SADE) Study Group of Endoscopic Retrograde Cholangio-Pancreaticography. Endoscopic classification of the papilla of Vater: Results of an inter- and intraobserver agreement study","volume":"5","author":"Haraldsson","year":"2017","journal-title":"United Eur. Gastroenterol. J."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Hager, P., Menten, M.J., and Rueckert, D. (2023). Best of both worlds: Multimodal contrastive learning with tabular and imaging data. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR52729.2023.02291"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"119505","DOI":"10.1016\/j.neuroimage.2022.119505","article-title":"Alzheimer\u2019s Disease Neuroimaging Initiative. DAFT: A universal module to inter-weave tabular data and 3D images in CNNs","volume":"260","author":"Wolf","year":"2022","journal-title":"NeuroImage"},{"key":"ref_11","first-page":"46","article-title":"Harnessing Effectiveness of ResNet-50 and EfficientNet for Few-Shot Learning","volume":"5","author":"Vajrangi","year":"2023","journal-title":"Int. J. Comput. Commun. Inform."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"101431","DOI":"10.1016\/j.eclinm.2022.101431","article-title":"A novel machine learning model and a public online prediction platform for prediction of post-ERCP-cholecystitis (PEC)","volume":"48","author":"Zhang","year":"2022","journal-title":"EClinicalMedicine"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1016\/j.gie.2024.08.009","article-title":"Development and validation of a machine learning\u2013based, point-of-care risk calculator for post-ERCP pancreatitis and prophylaxis selection","volume":"101","author":"Brenner","year":"2025","journal-title":"Gastrointest. Endosc."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"5107","DOI":"10.1007\/s00464-025-11937-5","article-title":"Machine learning-based predic-tion model for post-ERCP cholangitis in patients with malignant biliary obstruction: A retrospective multicenter study","volume":"39","author":"Jin","year":"2025","journal-title":"Surg. Endosc."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"e588","DOI":"10.1097\/MPA.0000000000002476","article-title":"Performance of clinical risk prediction models for post-ERCP pancreatitis: A systematic review","volume":"54","author":"Sabrie","year":"2025","journal-title":"Pancreas"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Andrzej, J. (2024). Impact of the shape of the ampulla of Vater and biliary tree pathology on access technique and post-ERCP complications. Res. Sq.","DOI":"10.21203\/rs.3.rs-4370114\/v1"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"7497","DOI":"10.1038\/s41598-018-25842-6","article-title":"Automatic anatomical classification of esophagogastroduodenoscopy images using deep convolutional neural networks","volume":"8","author":"Takiyama","year":"2018","journal-title":"Sci. Rep."},{"key":"ref_18","doi-asserted-by":"crossref","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":"ref_19","doi-asserted-by":"crossref","first-page":"7240129","DOI":"10.1155\/2020\/7240129","article-title":"Concrete cracks detection using convolutional neural network based on transfer learning","volume":"2020","author":"Su","year":"2020","journal-title":"Math. Probl. Eng."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"50755","DOI":"10.1109\/ACCESS.2021.3069022","article-title":"A classification of arab ethnicity based on face image using deep learning approach","volume":"9","author":"Prince","year":"2021","journal-title":"IEEE Access"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Gatys, L.A., Ecker, A.S., and Bethge, M. (2015). A neural algorithm of artistic style. arXiv.","DOI":"10.1167\/16.12.326"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"He, K., Gkioxari, G., Doll\u00e1r, P., and Girshick, R. (2017). Mask R-CNN. Proceedings of the IEEE International Conference on Computer Vision, IEEE.","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"384","DOI":"10.1016\/j.neunet.2020.03.023","article-title":"Automatic detection of tympanic membrane and middle ear infection from oto-endoscopic images via convolutional neural networks","volume":"126","author":"Khan","year":"2020","journal-title":"Neural Netw."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1562","DOI":"10.1109\/TMI.2018.2791721","article-title":"Interactive medical image segmentation using deep learning with image-specific fine tuning","volume":"37","author":"Wang","year":"2018","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Nie, D., Zhang, H., Adeli, E., Liu, L., and Shen, D. (2016). 3D deep learning for multi-modal imaging-guided survival time prediction of brain tumor patients. Medical Image Computing and Computer-Assisted Intervention\u2014MICCAI 2016, Springer.","DOI":"10.1007\/978-3-319-46723-8_25"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Fu, H., Xu, Y., Lin, S., Wong, D.W.K., and Liu, J. (2016). DeepVessel: Retinal vessel segmentation via deep learning and conditional random field. Medical Image Computing and Computer-Assisted Intervention\u2014MICCAI 2016, Springer.","DOI":"10.1007\/978-3-319-46723-8_16"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"1","DOI":"10.24018\/ejece.2022.6.5.458","article-title":"DenseNet based model for plant diseases diagnosis","volume":"6","author":"Bakr","year":"2022","journal-title":"Eur. J. Electr. Eng. Comput. Sci."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"8171","DOI":"10.1007\/s00464-025-12169-3","article-title":"Machine learning-derived predictive model for post-ERCP pancreatitis in patients with common bile duct stones: A retrospective multicenter study","volume":"39","author":"Chen","year":"2025","journal-title":"Surg. Endosc."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Cahyadi, O., Tehami, N., de-Madaria, E., and Siau, K. (2022). Post-ERCP pancreatitis: Prevention, diagnosis and management. Medicina, 58.","DOI":"10.3390\/medicina58091261"},{"key":"ref_30","unstructured":"Arabpour, E., Sadeghi, A., Rastegar, R., Mohammadi, P., Safavi-Naini, S.A., Moghadam, P.K., and Zali, M.R. (J. Hepato-Biliary-Pancreat. Sci., 2026). Predicting Post-ERCP Pancreatitis Using Machine Learning: Risk Stratification and Feature Importance Analysis, J. Hepato-Biliary-Pancreat. Sci., Epub ahead of printing."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Choi, Y., Choi, M., Kim, M., Ha, J., Kim, S., and Choo, J. (2018). StarGAN: Unified generative adversarial networks for multi-domain image-to-image translation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR.2018.00916"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Karras, T., Laine, S., and Aila, T. (2019). A style-based generator architecture for generative adversarial networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR.2019.00453"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/12\/1824\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T04:17:47Z","timestamp":1781842667000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/12\/1824"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,12]]},"references-count":32,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["diagnostics16121824"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16121824","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,12]]}}}