[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117117-en":3,"doc-seo-117117-105":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},117117,4398048949847,"Eliana","https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267",8,"Research & Report","Automated Quality Control by Application of Machine Learning Techniques for Quantitative Liver MRI","Quantitative magnetic resonance imaging (qMRI) and multi-parametric MRI are increasingly used to diagnose and monitor liver disease, yet these acquisitions are more complex than conventional T1- and T2-weighted scans and are more vulnerable to image-quality problems and artefacts. Reliable qMRI outputs require rigorous quality control, but manual QC is time-consuming, subjective, and prone to error. The thesis develops automated QC pipelines for liver qMRI using machine learning and deep learning, designed to be tag-free where possible for transfer across techniques, and evaluates performance with added metadata and external pilot settings.","Automated Quality Control by Application of Machine Learning Techniques for Quantitative  \nLiver MRI  \nCharles Hill  \nSt Hugh's College  \nUniversity of Oxford A thesis presented for the degree of Doctor of Philosophy  \nMichaelmas 2021  \nCopyright © 2021 by Charles Hill All Rights Reserved  \nAcknowledgements  \nThis thesis wouldn't have been possible without the support of many people, both inside and outside academia. First of all, I would like to thank my three supervisors; Prof Vicente Grau, Dr Luca Biasiolli, and Prof Matthew Robson. Vicente and the group have been invaluable in guiding me into the world of deep learning and being around for a cycle and pint. Luca has somehow been able to support me all the way from Italy while supporting his own young family, and starting a new job. Finally, I am grateful to Matt for probably being the most supportive industry supervisor that existed. Not only did he actually meet me regularly, but he also had some good ideas.  \nI would also like to thank those who I have lived with, over the years in Oxford. The 50 Windmill Road collective, and the Reliance Way Massive have reminded me that life isn't always about work. Many great memories have been made over the years which I will remember fondly. Thanks also go out to John McGonigleand his team at Perspectum.  \nFinally I would like to express thanks to family, and my partner for their invaluable support over the past four years. They have been there consistently, through thick and thin. I wouldn't have finished this thesis without them.  \nDeclaration  \nI, Charles E. Hill, declare that this thesis is entirely my own work, and except where otherwise stated, describes my own research.  \nCharles Hill 13/02/2024  \nC Hill 10:21, Feb 13, 2024 UTC  \nSignature  \nDate  \nAbstract  \nQuantitative magnetic resonance imaging (qMRI) and multi-parametric MRI are being increasingly used to diagnose and monitor liver diseases such as non-alcoholic fatty liver disease (NAFLD) . These acquisitions are comparably more complicated than traditional T1-weighted and T2-weighted MRI scans and are also more prone to image quality issues and artefacts. In order for the output of the qMRI scans to be useable, they must undergo a rigorous and often lengthy quality control (QC) . This manual QC is prone to human error and subjective. Additionally, with the development of new qMRI techniques, this leads to the manifestation of new quality issues. This thesis focuses on the development and implementation of automated QC processes for liver qMRI scans, that is where possible tag-free such that the process can be adapted to different imaging techniques. These automated QC processes were implemented using a variety of machine learning (ML) and deep learning (DL) approaches. These methods, developed on T1 mapping in UKBiobank, were designed to output metrics from the MRI scans that could be used to identify a specific quality issue, such as in chapter 3, or give amore general indication of the image quality in chapter 4 . Furthermore, it was hypothesised that the introduction of associated meta-data, such as patient factors and scanning parameters, into these deep learning models would increase overall performance. This was explored in chapter 5. Finally, in order to assess the utility of our developed algorithms in a wider setting except for T1 mapping in UKBiobank, we tested it in two settings. Pilot study one assessed the utility of the model in T1 mapping in a separate study (CoverScan) . Pilot study two assessed the utility of the model in a different qMRI acquisition; proton density fat fraction (PDFF) acquisitions from UKBiobank.  \nKeywords—Liver MRI-Deep Learning-Quality Control  \nTable of Contents  \n1 Introduction 2  \n1.1 Motivation .............................. 2  \n1.2 Objectives .............................. 3  \n1.3 Thesis Outline ............................ 5  \n1.4 Published Work ........................... 6  \n1.4.1 Conference Abstracts .................... 6  ","cbCaieArUkwwWDzS","https://ap.wps.com/l/cbCaieArUkwwWDzS","pdf",19693330,1,182,"English","en",105,"# Introduction\n## Motivation\n## Objectives\n## Thesis Outline\n## Published Work\n# Background\n## Liver Disease and Diagnosis\n## Magnetic Resonance Imaging\n## Quantitative MRI\n## Multi-parametric Liver MRI\n## Artefacts in MRI\n## Image Quality Assessment (IQA)\n## Machine and Deep Learning\n# Landmark Detection for Slice Location identification\n## Introduction\n## Methods\n## Experiments\n## Results\n## Discussion\n## Conclusion\n# Data imputation for detection","[{\"question\":\"Why is automated quality control needed for quantitative liver MRI?\",\"answer\":\"qMRI acquisitions are more complex than traditional MRI and are prone to quality issues and artefacts. Manual QC is lengthy, subjective, and error-prone, so automated QC improves reliability for usable outputs.\"},{\"question\":\"How do the proposed automated QC methods work?\",\"answer\":\"The thesis develops automated QC processes for liver qMRI using machine learning and deep learning models. The methods are trained to output metrics that identify specific quality issues or provide a general indication of image quality.\"},{\"question\":\"What role does metadata play in improving model performance?\",\"answer\":\"The thesis investigates whether including associated metadata—such as patient factors and scanning parameters—improves overall deep learning performance.\"}]","Automated Quality Control by Application of Machine Learning Techniques for Quantitative Liver MRI | PDF",1785673973,459,{"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},"automated-quality-control-by-application-of-machine-learning-techniques-for-quantitative-liver-mri","",{"@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/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/automated-quality-control-by-application-of-machine-learning-techniques-for-quantitative-liver-mri/117117/",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},"Why is automated quality control needed for quantitative liver MRI?","Question",{"text":75,"@type":76},"qMRI acquisitions are more complex than traditional MRI and are prone to quality issues and artefacts. Manual QC is lengthy, subjective, and error-prone, so automated QC improves reliability for usable outputs.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How do the proposed automated QC methods work?",{"text":80,"@type":76},"The thesis develops automated QC processes for liver qMRI using machine learning and deep learning models. The methods are trained to output metrics that identify specific quality issues or provide a general indication of image quality.",{"name":82,"@type":73,"acceptedAnswer":83},"What role does metadata play in improving model performance?",{"text":84,"@type":76},"The thesis investigates whether including associated metadata—such as patient factors and scanning parameters—improves overall deep learning performance.","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},[93,97,101,105,110,115,120,123,128,131,135],{"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":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},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"]