[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126222-en":3,"doc-seo-126222-105":31,"detail-sidebar-cat-0-en-105":93},{"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},126222,549768702563,"Sage","https://ap-avatar.wpscdn.com/avatar/8000c4aa63b76e948b?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786536092046926083",8,"Research & Report","Machine Learning Methods for Phenotyping in Drug Discovery - Doctor of Philosophy Thesis","The surge in Electronic Health Record (EHR) availability creates new opportunities and challenges for healthcare research. Phenotyping—the identification of observable traits linked to genetic or environmental variation—is essential for disease understanding and effective drug discovery. As EHR complexity increases, conventional approaches often struggle. This thesis applies machine learning to extract more accurate EHR patterns, improving phenotyping precision and breadth. It develops methods for temporal modelling, addresses label noise, refines patient subtyping, and advances survival analysis for better representations and downstream predictions.","Machine Learning Methods for Phenotyping in Drug Discovery  \nAndre Vauvelle  \nA dissertation submitted in partial fulfillment of the requirements for the degree of  \nDoctor of Philosophy  \nof  \nUniversity College London.  \nInstitute of Health Informatics  \nUniversity College London  \nSeptember 22, 2024  \n2  \nI, Andre Vauvelle, confirm that the work presented in this thesis is my own. Where information has been derived from other sources, I confirm that this has been indicated in the work.  \nAbstract  \nThe surge in availability of Electronic Health Record (EHR) data presents both opportunities and challenges for healthcare research. Effective phenotyping, the process of identifying observable traits related to genetic or environmental variations, is integral to understanding diseases and advancing drug discovery. With the increasing complexity of EHR data, traditional methods often fall short. Machine learning, due to its adaptive and data-driven nature, holds the promise of more accurately deciphering patterns within EHR, thus enhancing the precision and scope of phenotyping.  \nWithin this work, we address a broad spectrum of challenges associated with structured EHR data and its practical applications, highlighting our efforts to enhance phenotyping in the context of drug discovery. The initial research explores EHR’s temporal dynamics, leveraging the signature transform as an alternative modelling paradigm for sequential data in heart failure prediction models. Subsequent investigations confront the pervasive issue of label noise in EHR data. By integrating positive and unlabelled learning with transformer architectures, we attempt to enrich cohorts by identifying missed diagnoses and recover genomic associations with greater power. Moreover, the work navigates the realm of patient subtyping, weighing the merits of reconstruction against outcome objectives to forge accurate EHR data representations. In the final chapter of this research, we advance the field of survival analysis by integrating differentiable sorting methods with partial order supervision. The method serves as an alternative to the conventional Cox’s partial likelihood with the advantage of a transitive inductive prior.  \nAcknowledgements  \nFirst and foremost, I would like to express my deepest gratitude to Benevolent AI for sponsoring my research. Special thanks go to Dr. P´aid´ı Creed, Dr. Aaron Sim, Dr. Hamish Tomlinson, Dr. Aylin Cakiroglu, Anna Mu˜noz Farr´e, Antonios Poulakakis Daktylidis, and the entire team at Benevolent AI for their invaluable assistance and insights.  \nI would also like to extend my appreciation to my fellow CDT students, especially Joe Farringdon, who has been a constant source of support and a patient sounding board for my sometimes unconventional ideas.  \nTo my esteemed supervisors, Prof. Spiros Denaxas and Dr. Nicola Richmond, thank you for your unwavering guidance and for allowing me the freedom to explore new avenues in my research. Special recognition must go to Dr. Benjamin Wild, who has been an invaluable mentor in the development of the Diffsurv chapter.  \nI am also deeply grateful to my friends and family who have supported me throughout this journey. Endless thanks to Katarzyna Bruzda, for holding the mirror that reflected my better self when all I could see were my flaws.  \nTo all of you,頑張って(Ganbatte)!  \nOriginality  \nStatement: The work presented in Chapters 2, 4, and 5 of this thesis has been accepted for publication, for which I am the lead author. Text from these papers has been incorporated into the respective chapters, and may include minor edits contributed by collaborators. Chapter 3 was completed as part of an internship at Benevolent AI and had significant input from Anna Mu˜noz Farr´e and Antonios Poulakakis Daktylidis. Below is a list of chapters and their respective publication.  \nChapter 2: Andre Vauvelle, Paidi Creed, and Spiros Denaxas. Neuralsignature methods for structured EHR prediction. BMC Medical ","cbCaiboax6PI7BZf","https://ap.wps.com/l/cbCaiboax6PI7BZf","pdf",5679271,5,1,157,"English","en",105,"# Contents\n## Introduction\n### Aims and Objectives\n### Thesis Summary\n### Non-Technical Summary\n## Neural-Signature Methods for Structured EHR Prediction\n## Phenotyping with Positive Unlabelled Learning for Genome-Wide Association Studies\n## (Additional chapters continue)","[{\"question\":\"Why is phenotyping important for drug discovery in this thesis?\",\"answer\":\"Phenotyping links observable traits to genetic or environmental variation, supporting disease understanding and improving the development of drug discovery strategies.\"},{\"question\":\"How does the thesis improve phenotyping when traditional methods are insufficient?\",\"answer\":\"It uses machine learning to better capture complex patterns in EHR data, including temporal dynamics and data-driven modelling approaches.\"},{\"question\":\"What key challenges in EHR data does the thesis address?\",\"answer\":\"It targets structured EHR temporal modelling, pervasive label noise, patient subtyping representation learning, and survival analysis improvements using differentiable sorting with partial order supervision.\"}]","Machine Learning Methods for Phenotyping in Drug Discovery - Doctor of Philosophy Thesis | PDF",1785903896,396,{"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":88,"head_meta":90,"extra_data":92,"updated_unix":29},"machine-learning-methods-for-phenotyping-in-drug-discovery-doctor-of-philosophy-thesis","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/machine-learning-methods-for-phenotyping-in-drug-discovery-doctor-of-philosophy-thesis/126222/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"Why is phenotyping important for drug discovery in this thesis?","Question",{"text":77,"@type":78},"Phenotyping links observable traits to genetic or environmental variation, supporting disease understanding and improving the development of drug discovery strategies.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How does the thesis improve phenotyping when traditional methods are insufficient?",{"text":82,"@type":78},"It uses machine learning to better capture complex patterns in EHR data, including temporal dynamics and data-driven modelling approaches.",{"name":84,"@type":75,"acceptedAnswer":85},"What key challenges in EHR data does the thesis address?",{"text":86,"@type":78},"It targets structured EHR temporal modelling, pervasive label noise, patient subtyping representation learning, and survival analysis improvements using differentiable sorting with partial order supervision.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},"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":20,"slug":139},19,"General","general"]