[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119760-en":3,"doc-seo-119760-105":29,"detail-sidebar-cat-0-en-105":90},{"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":11,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":26,"seo_description":14,"update_tm":27,"read_time":28},119760,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","Machine learning for mortality risk prediction with changing patient demographics","Over the last 25–30 years, intensive care unit (ICU) mortality risk prediction models have advanced considerably, with the ICNARC model (2007) widely used due to strong performance on UK data from over 230,000 admissions. Existing models, however, degrade when patient cohort demographics shift, such as population aging, and typically require periodic recalibration. The paper proposes a machine learning pipeline for training mortality prediction models and extends it to continuous online retraining, improving performance across varying demographic scenarios.","Machine learning for mortality risk prediction with changing patient demographics  \nWAINWRIGHT, Richard and SHENFIELD, Alex \u003C [http://orcid.org/0000-0002-](http://orcid.org/0000-0002-)[ ](http://orcid.org/0000-0002-)[2931-8077](2931-8077)>  \nAvailable from Sheffield Hallam University Research Archive (SHURA) at: [http://shura.shu.ac.uk/32289/](http://shura.shu.ac.uk/32289/)  \nThis document is the author deposited version. You are advised to consult the publisher's version if you wish to cite from it.  \nPublished version  \nWAINWRIGHT, Richard and SHENFIELD, Alex (2023) . Machine learning for mortality risk prediction with changing patient demographics. In: 2023 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB) . IEEE.  \nCopyright and re-use policy  \nSee [http://shura.shu.ac.uk/information.html](http://shura.shu.ac.uk/information.html)  \nSheffield Hallam University Research Archive  \n[http://shura.shu.ac.uk](http://shura.shu.ac.uk)  \nMachine learning for mortality risk prediction with changing patient demographics  \nRichard Wainwright  \nDepartment of Engineering and Mathematics Sheffield Hallam University Sheffield, UK [ric.wainwright@theinsightsfamily.com](ric.wainwright@theinsightsfamily.com)  \nAlex Shenfield  \nDepartment of Engineering and Mathematics Sheffield Hallam University Sheffield, UK[a.shenfield@shu.ac.uk](a.shenfield@shu.ac.uk)  \nAbstract—Over the last 25-30 years there has been significant work carried out in producing risk prediction models for patients admitted to intensive care units. The most recent of these models in widespread use is the Intensive Care National Audit and Research Centre (ICNARC) model developed in 2007 which uses data from more than 230,000 admissions to UK intensive care units to develop and validate a UK based model outperforming other approaches. However, as with the majority of risk prediction models, the ICNARC model struggles with changing patient cohort demographics (such as the aging populations seen currently in the western world) and requires periodic recalibration.  \nThis paper introduces a machine learning pipeline for developing mortality prediction models and uses it to train a variety of ML models. The top performing of these outperform current commonly used mortality risk prediction models such as APACHE-II, SAPS-II, and the ICNARC model. This machine learning pipeline is then extended to allow continuous retraining via online learning. The results show that it is possible to retrain our model at different intervals to deal with varying patient demographics -improving model performance across a range of different patient cohort scenarios.  \nIndex Terms—Machine Learning, Online Learning, Risk Prediction, Mortality Prediction, ICU  \nI. INTRODUCTION  \nOver the last decade, UK National Health Service (NHS) waiting times have increasingly failed to meet targets. This is now compounded by challenges in recruitment and retention of nursing and other hospital staff meaning that intensive care unit (ICU) staffing ratios are often well below recommended levels [1] . To mitigate the impact of this staffing crisis, prediction of mortality risk in the ICU environment is of increasing importance to allow stratification of patients and targeted support.  \nThere has been significant work carried out in developing risk prediction models for patients admitted to intensive care units [2], [3] . In the UK, the most recent of these is the Intensive Care National Audit and Research Centre (ICNARC) model which uses data from 163 intensive care units across the UK (covering 231,900 admissions) to develop and validate a model outperforming previous approaches such as the SAPSII (Simplified Acute Physiology Score) [4] and APACHE-II (Acute Physiology and Chronic Evaluation) [5] scores.  \nHowever, a key disadvantage of these mortality risk prediction models is that they are developed using retrospective pa-  \ntient cohorts, with patient demographics and the","cbCaihFMf1nfqfL8","https://ap.wps.com/l/cbCaihFMf1nfqfL8","pdf",3845465,1,"English","en",105,"# Introduction\n## Related work\n# Machine learning pipeline and online retraining\n# Dataset and methodology\n# Experimental results\n# Conclusions and future work","[{\"question\":\"Why do existing ICU mortality risk models struggle over time?\",\"answer\":\"They are often built on retrospective patient cohorts, while demographics and treatment effectiveness change. This causes performance drift and can require periodic recalibration.\"},{\"question\":\"What approach does the paper introduce for mortality risk prediction?\",\"answer\":\"It presents a machine learning pipeline to develop mortality prediction models, training multiple ML approaches and identifying top performers against established baselines.\"},{\"question\":\"How does the paper handle changing patient demographics after training?\",\"answer\":\"It extends the pipeline with continuous retraining via online learning, allowing the model to adapt at different intervals to demographic changes and maintain performance.\"}]","Machine learning for mortality risk prediction with changing patient demographics | PDF",1785726163,20,{"code":4,"msg":30,"data":31},"ok",{"site_id":23,"language":22,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"machine-learning-for-mortality-risk-prediction-with-changing-patient-demographics","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/machine-learning-for-mortality-risk-prediction-with-changing-patient-demographics/119760/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":22,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Why do existing ICU mortality risk models struggle over time?","Question",{"text":74,"@type":75},"They are often built on retrospective patient cohorts, while demographics and treatment effectiveness change. This causes performance drift and can require periodic recalibration.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What approach does the paper introduce for mortality risk prediction?",{"text":79,"@type":75},"It presents a machine learning pipeline to develop mortality prediction models, training multiple ML approaches and identifying top performers against established baselines.",{"name":81,"@type":72,"acceptedAnswer":82},"How does the paper handle changing patient demographics after training?",{"text":83,"@type":75},"It extends the pipeline with continuous retraining via online learning, allowing the model to adapt at different intervals to demographic changes and maintain performance.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":28,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":28,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]