[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117045-en":3,"doc-seo-117045-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},117045,1649267921044,"Ava Thompson","https://us-avatar.wpscdn.com/avatar/1800007509477c92dfb?_k=1782875107921204101",7,"Healthcare","Complication Prediction after Esophagectomy with Machine Learning - Academic Article","Esophageal cancer is treated effectively with esophagectomy, yet postoperative complications remain frequent. This study evaluates how machine learning can predict anastomotic leakage and pneumonia up to two days in advance using a multimodal temporal dataset from 417 patients (2011–2021). Inputs include laboratory results, vital signs, thorax images, and preoperative characteristics. The best models achieve mean AUROCs of 0.87 and 0.82 for leakage (1 and 2 days ahead) and 0.74 and 0.61 for pneumonia. Results support machine-learning-based clinical decision support after surgery.","diagnostics  \nArticle  \nComplication Prediction after Esophagectomy with Machine Learning  \nJorn-Jan van de Beld 1,2,*, David Crull 2, Julia Mikhal 2,3, Jeroen Geerdink 2, Anouk Veldhuis 2, Mannes Poel 1 and Ewout A. Kouwenhoven 2  \nCitation: van de Beld, J.-J.; Crull, D.; Mikhal, J.; Geerdink, J.; Veldhuis, A.; Poel, M.; Kouwenhoven, E.A. Complication Prediction after Esophagectomy with Machine Learning. Diagnostics 2024, 14, 439 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)diagnostics14040439  \nAcademic Editor: Yanwu Xu  \nReceived: 21 November 2023  \nRevised: 28 December 2023  \nAccepted: 29 December 2023  \nPublished: 17 February 2024  \nCopyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) .  \n1 Faculty of EEMCS, University of Twente, 7500 AE Enschede, The Netherlands  \n2 Hospital Group Twente (ZGT), 7609 PP Almelo, The Netherlands  \n3 Faculty of BMS, University of Twente, 7500 AE Enschede, The Netherlands  \n* [Correspondence: j.j.vandebeld@utwente.nl](Correspondence: j.j.vandebeld@utwente.nl)  \nAbstract: Esophageal cancer can be treated effectively with esophagectomy; however, the postoperative complication rate is high. In this paper, we study to what extent machine learning methods can predict anastomotic leakage and pneumonia up to two days in advance. We use a dataset with 417 patients who underwent esophagectomy between 2011 and 2021 . The dataset contains multimodal temporal information, specifically, laboratory results, vital signs, thorax images, and preoperative patient characteristics. The best models scored mean test set AUROCs of 0.87 and 0.82 for leakage 1 and 2 days ahead, respectively. For pneumonia, this was 0.74 and 0.61 for 1 and 2 days ahead, respectively. We conclude that machine learning models can effectively predict anastomotic leakage and pneumonia after esophagectomy.  \nKeywords: esophagectomy; clinical decision support; multimodal machine learning; temporal learning  \n1. Introduction  \nAccording to the World Cancer statistics in 2020, esophageal cancer ranks seventh in terms of incidence and sixth in mortality overall. The disease is more common in men (70%) and most prevalent in eastern Asia [1] .  \nIn the past decade, minimally invasive robot-assisted techniques have become increasingly popular as an alternative to open esophagectomy. Clinical trials have shown that minimally invasive procedures lower the risk of postoperative complications, specifically, pulmonary complications [2,3] . Nevertheless, postoperative complications are common, with a rate of 65% reported by a nationwide study in the Netherlands including 1617 patients [4] . In this study, pneumonia (21%) and anastomotic leakage (AL) (19%) were the most common postoperative complications. A similar study with 2704 patients from centers across 14 different countries found a postoperative complication rate of 59% and rates of 14.6% and 11.4% for pneumonia and AL, respectively [5] .  \nIn recent years, there have been considerable advances in the field of medical artificial intelligence (MAI) with successes in a wide range of retrospective studies [6] . For example, pneumonia detection models have been developed, with most reporting an accuracy over 90%; logistic regression (LR) and deep learning (DL) models are most commonly used for this task [7] . Yet, great challenges remain in the field of MAI, for example, the development of multimodal models that can handle various medical data sources as input [6] .  \nMachine learning methods have been employed to analyze and predict complications post-esophagectomy. Early studies, often limited to preoperative variables, focused on the identification of preoperative ris","cbCaie04zBpnMkRk","https://ap.wps.com/l/cbCaie04zBpnMkRk","pdf",868957,1,14,"English","en",105,"# Introduction\n## Epidemiology and postoperative complication burden\n## Medical AI and multimodal modeling challenges\n# Materials and Methods\n## Data sources and dataset description","[{\"question\":\"What complications does the machine learning approach aim to predict after esophagectomy?\",\"answer\":\"It focuses on two common infectious complications: anastomotic leakage and pneumonia.\"},{\"question\":\"What time horizon does the study evaluate for prediction?\",\"answer\":\"Prediction targets anastomotic leakage and pneumonia up to two days in advance.\"},{\"question\":\"Which types of data are used as model inputs?\",\"answer\":\"The dataset includes preoperative patient characteristics, laboratory results, vital signs, and thorax radiology images, organized as multimodal temporal information.\"}]","Complication Prediction after Esophagectomy with Machine Learning - Academic Article | PDF",1785673340,35,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"complication-prediction-after-esophagectomy-with-machine-learning-academic-article","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/healthcare/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/complication-prediction-after-esophagectomy-with-machine-learning-academic-article/117045/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What complications does the machine learning approach aim to predict after esophagectomy?","Question",{"text":75,"@type":76},"It focuses on two common infectious complications: anastomotic leakage and pneumonia.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What time horizon does the study evaluate for prediction?",{"text":80,"@type":76},"Prediction targets anastomotic leakage and pneumonia up to two days in advance.",{"name":82,"@type":73,"acceptedAnswer":83},"Which types of data are used as model inputs?",{"text":84,"@type":76},"The dataset includes preoperative patient characteristics, laboratory results, vital signs, and thorax radiology images, organized as multimodal temporal information.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,118,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":116,"slug":117},40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]