[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117227-en":3,"doc-seo-117227-105":31,"detail-sidebar-cat-0-en-105":92},{"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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},117227,2336464648322,"Aria","https://ap-avatar.wpscdn.com/avatar/2200025388227c56fec?_k=1778556882303663488",8,"Research & Report","Machine Learning Methods for Tabular Electronic Health Record Data - Doctor of Philosophy Thesis","Electronic Health Record (EHR) data represents demographics and discrete events across a patient’s life, yet it is often difficult for machine learning due to manual entry errors, missing values, multiple time-series with widely varying granularity, high dimensionality, and limited patient counts. This thesis compiles theoretical and empirical work on methods tailored to EHR settings. It evaluates recurrent neural networks and transformer models, proposes novel transformer architectures, and reports accuracy above 94% across evaluated tasks.","Machine Learning Methods for Tabular Electronic Health Record Data  \nOdhran O’Donoghue  \nGreen Templeton College  \nUniversity of Oxford  \nA thesis submitted for the degree of Doctor of Philosophy  \nMichaelmas 2023  \nAbstract  \nThe vast majority of the world’s medical data exists in the Electronic Health Record (EHR), a document used to record demographic characteristics and discrete events throughout a patient’s life. This EHR data is often poorly suited for machine learning, consisting of manually recorded data prone to errors and missing values, multiple time-series at highly varying levels of granularity, and typically high dimensionality and low patient numbers in datasets.  \nNevertheless, machine learning on EHR data has a high potential for impact, and it is necessary to evaluate the efficacy of different machine learning methods for medical data. This thesis is an accumulation of theoretical and empirical work on developing machine learning methods appropriate for EHR data. Three work components are described.  \nThe first work component of this thesis is a systematic evaluation of recurrent neural networks and transformer models for machine learning on EHR data. This thesis also introduces novel transformer-based architectures with properties that are desirable for EHR data. Transformers are identified as models with high performance for this task, achieving accuracy scores exceeding 94% in all evaluated tasks.  \nThe second work component of this thesis is an effort to build machine models that can predict COVID-19 vaccine adverse events with a less than 0.1% occurrence rate in training data. An Area Under Receiver Operating Curve (AUROC) of 70.1% was achieved on this task, and feature and subgroup analysis to inform clinical understanding was performed.  \nThe third work component is an effort to build machine learning models to predict stroke deterioration events from small neurointensive care EHR datasets. We evaluate a range of machine learning architectures, and we identify architectures that perform well both on training data and external validation data. We perform further feature and subgroup analysis to help place this work in the wider context of our clinical understanding of ischaemic stroke and malignant cerebral oedema.  \nThis thesis ultimately finds that a diverse range of models are needed for EHR data in different contexts - a ’one size fits all’ approach to machine learning is insufficient to address the diverse range of tasks, requirements and data formats within this setting.  \nMachine Learning Methods for Tabular Electronic Health Record Data  \nOdhran O’Donoghue Green Templeton College  \nUniversity of Oxford  \nA thesis submitted for the degree of Doctor of Philosophy  \nMichaelmas 2023  \niv  \nThis thesis is dedicated to Helen Feazey and Michael O’Donoghue. I know how rare it is to have truly unconditional love, and I will never once take that for granted. Thank you for being lights when all other lights went out.  \nThank you for the chances you gave me.  \nDad.  \nMum.  \nI love you both more than anything else.  \nPrinted on October 5, 2024  \nv  \nPrinted on October 5, 2024  \nAcknowledgements  \nI would be nowhere without my supervisor Professor David Clifton. Thank you for being the intellectual backbone upon which my work has rested, and being a constant sounding board for my academic trajectory. These thanks extend to the rest of the people I have collaborated with in the Computational Health Informatics group-Andrew Soltan, Jeny Yang, and in particular Anshul Thakur. Although our closeness may be based on a shared trauma of using Windows remote desktop for academic work, it was phenomenal to conduct research with you.  \nI would also be nowhere without my collaborations with Professor Agni Orfanoudaki at the Said Business School, Professor Charlene Ong and Professor Stelios Smirnakis from Massachusetts General Hospital, and Panos Tsimpos from MIT. Working with you all has been the highlight of my PhD. I ","cbCainLcToL3vYwI","https://ap.wps.com/l/cbCainLcToL3vYwI","pdf",14006466,2,1,233,"English","en",105,"# Abstract\n# Thesis Overview\n## Recurrent Neural Networks and Transformers\n## COVID-19 Vaccine Adverse Event Prediction\n## Stroke Deterioration Prediction from Neurointensive Care Data\n## Key Conclusion on Model Diversity","[{\"question\":\"Why is EHR data challenging for machine learning?\",\"answer\":\"EHR datasets often contain manually recorded values prone to errors and missingness, multiple time-series at different granularities, high dimensionality, and relatively small numbers of patients.\"},{\"question\":\"What modeling families does the thesis evaluate for EHR machine learning?\",\"answer\":\"The thesis systematically evaluates recurrent neural networks and transformer models, and it introduces transformer-based architectures designed to suit EHR data.\"},{\"question\":\"How does the thesis conclude regarding applying a single model across contexts?\",\"answer\":\"The thesis concludes that different EHR tasks, requirements, and data formats require a diverse set of models, and a one-size-fits-all approach is insufficient.\"}]","Machine Learning Methods for Tabular Electronic Health Record Data - Doctor of Philosophy Thesis | PDF",1785674504,587,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"machine-learning-methods-for-tabular-electronic-health-record-data-doctor-of-philosophy-thesis","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,48,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":20},"https://docshare.wps.com/document/","Document",{"item":49,"name":12,"@type":44,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/machine-learning-methods-for-tabular-electronic-health-record-data-doctor-of-philosophy-thesis/117227/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-02",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Why is EHR data challenging for machine learning?","Question",{"text":76,"@type":77},"EHR datasets often contain manually recorded values prone to errors and missingness, multiple time-series at different granularities, high dimensionality, and relatively small numbers of patients.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What modeling families does the thesis evaluate for EHR machine learning?",{"text":81,"@type":77},"The thesis systematically evaluates recurrent neural networks and transformer models, and it introduces transformer-based architectures designed to suit EHR data.",{"name":83,"@type":74,"acceptedAnswer":84},"How does the thesis conclude regarding applying a single model across contexts?",{"text":85,"@type":77},"The thesis concludes that different EHR tasks, requirements, and data formats require a diverse set of models, and a one-size-fits-all approach is insufficient.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]