[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121135-en":3,"doc-seo-121135-105":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":20,"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},121135,8796095461610,"Oliver","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Clinical translation of machine learning algorithms for seizure detection in scalp electroencephalography - systematic review","Machine learning algorithms for seizure detection show strong diagnostic promise, including very high reported accuracies, yet few published methods satisfy the practical requirements for successful clinical translation. Key barriers include limited generalisability from training data, performance variation across electroencephalography acquisition hardware, and runtime processing costs that may hinder real-time clinical deployment. A systematic review evaluates translatability using criteria such as generalisability, run-time cost, explainability, and clinically relevant performance metrics, while also supporting non-specialists with domain context for model development and evaluation.","arXiv :2404 . 15332v2 [ ee ss . SP] 13 Aug 2024  \nClinical translation of machine learning algorithms for seizure detection in scalp electroencephalography:  \nsystematic review  \nNina Moutonnet 1 , Steven White3 , Benjamin P Campbell4 , Saeid Sanei5 , Toshihisa Tanaka6 , Hong Ji7 , Danilo Mandic5,8 , and Gregory Scott2,8  \n1 Department of Computing, Imperial College London  \n2 Department of Brain Sciences, Imperial College London  \n3 Department of Clinical Neurophysiology, National Hospital for Neurology & Neurosurgery, London, United Kingdom  \n4 Department of Bioengineering, Imperial College London  \n5 Department of Electrical and Electronic Engineering, Imperial College London  \n6 Tokyo University of Agriculture and Technology  \n7 Shaanxi Provincial Key Laboratory of Fashion Design Intelligence, Xi’an  \nPolytechnic University  \n8 UK Dementia Research Institute  \nAbstract  \nMachine learning algorithms for seizure detection have shown considerable diagnostic potential, with recent reported accuracies reaching 100% . Yet, only few published algorithms have fully addressed the requirements for successful clinical translation. This is, for example, because the properties of training data may limit the generalisability of algorithms, algorithm performance may vary depending on which electroencephalogram (EEG) acquisition hardware was used, or run-time processing costs may be prohibitive to real-time clinical use cases. To address these issues in a critical manner, we systematically review machine learning algorithms for seizure detection with a focus on clinical translatability, assessed by criteria including generalisability, run-time costs, explainability, and clinically-relevant performance metrics. For non-specialists, the domain-specific knowledge necessary to contextualise model development and evaluation is provided. It is our hope that such critical evaluation of machine learning algorithms with respect to their potential realworld effectiveness can help accelerate clinical translation and identify gaps in the current seizure detection literature.  \nCorresponding author  \nDr Gregory Scott BEng MSc MBBS MRCP PhD Post-Doctoral, Post-CCT Research Fellow Honorary Consultant Neurologist UK DRI Care Research and Technology Centre 9th Floor, Sir Michael Uren Hub, Imperial College London, 86 Wood Lane, London. W12 0BZ. UK. Tel: +44 (0)7909  \n691484 Email: [gregory.scott99@imperial.ac.uk](gregory.scott99@imperial.ac.uk)  \nKeywords  \nclinical translation; deep learning; electroencephalography; epilepsy; machine learning; scalp; seizure detection  \n1 Introduction  \nA seizure is an abnormal synchronous excitation of one or more populations of neurons in the brain. Seizures are not a rare phenomenon, with an average lifetime incidence of 2-5%[1] . Seizures may occur in a range of clinical contexts, in particular in patients with epilepsy-defined as a tendency to unprovoked seizures [2] . Other, provoking, causes of seizures include central nervous system (CNS) infections, metabolic abnormalities, traumatic brain injuries, and drug toxicity. Seizure symptoms vary widely, from abnormal sensations to convulsions and altered awareness. Seizures are usually self-limiting, in that they typically last less than two minutes. Occasionally, they may continue for more than five minutes and/or recur without full recovery, a state termed status epilepticus, a medical emergency [3] .  \nThe accurate and timely detection of seizures is an important healthcare challenge [4] . Seizure detection is straightforward when the seizure activity has a well-recognised clinical correlate like generalised convulsions or a subjective alteration in experience, termed an aura. However, seizures without clear symptoms may be missed or mistaken for other medical phenomena. For example, ongoing seizure activity without an obvious motor component, known as non-convulsive status epilepticus (NCSE), affects 8-20% of patients in intensive care units (ICUs) and can be fatal","cbCaikh0bg2iXQZF","https://ap.wps.com/l/cbCaikh0bg2iXQZF","pdf",8464390,1,60,"English","en",105,"# Introduction\n## Clinical challenge of seizure detection\n## Seizure types and clinical context\n## Why automated seizure detection is needed\n## Scalp EEG and recording modalities\n# Systematic review focus and evaluation criteria\n## Generalisability and clinical transfer\n## Run-time costs and real-time use\n## Explainability and performance metrics","[{\"question\":\"Why is clinical translation difficult for seizure detection algorithms?\",\"answer\":\"Training data properties can limit generalisability, algorithm performance may change with EEG acquisition hardware, and runtime costs may be too high for real-time clinical use cases.\"},{\"question\":\"What criteria does the systematic review use to assess clinical translatability?\",\"answer\":\"Translatability is evaluated with criteria including generalisability, run-time costs, explainability, and clinically relevant performance metrics.\"},{\"question\":\"Why does the review focus on scalp electroencephalography (EEG)?\",\"answer\":\"Scalp EEG is non-invasive and contains subtle electrical abnormalities even when clear clinical correlates are absent, making it especially informative for automated seizure detection.\"}]","Clinical translation of machine learning algorithms for seizure detection in scalp electroencephalography - systematic review | PDF",1785734003,151,{"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},"clinical-translation-of-machine-learning-algorithms-for-seizure-detection-in-scalp-electroencephalography-systematic-review","",{"@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/clinical-translation-of-machine-learning-algorithms-for-seizure-detection-in-scalp-electroencephalography-systematic-review/121135/",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","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is clinical translation difficult for seizure detection algorithms?","Question",{"text":75,"@type":76},"Training data properties can limit generalisability, algorithm performance may change with EEG acquisition hardware, and runtime costs may be too high for real-time clinical use cases.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What criteria does the systematic review use to assess clinical translatability?",{"text":80,"@type":76},"Translatability is evaluated with criteria including generalisability, run-time costs, explainability, and clinically relevant performance metrics.",{"name":82,"@type":73,"acceptedAnswer":83},"Why does the review focus on scalp electroencephalography (EEG)?",{"text":84,"@type":76},"Scalp EEG is non-invasive and contains subtle electrical abnormalities even when clear clinical correlates are absent, making it especially informative for automated seizure detection.","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},[93,97,101,105,109,114,119,122,127,130,134],{"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":21,"slug":108},5,"Comic","comic",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]