[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121909-en":3,"doc-seo-121909-105":30,"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":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},121909,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Uncertainty Quantification in Machine Learning for Biosignal Applications - A Review","Uncertainty Quantification (UQ) improves the interpretability and robustness of machine learning predictions, especially for medical biosignals. EEG, ECG, EOG, and EMG are affected by low signal-to-noise ratios, and clinical use requires uncertainty estimates that support human interpretation. The review surveys state-of-the-art UQ methods for biosignal ML, detailing approaches, shortcomings, uncertainty measures, and theoretical frameworks. It highlights misconceptions, outlines recommendations and literature gaps, and discusses needs for diagnostic implementation and for prosthesis or brain-computer interface uncertainty handling, concluding that interaction design remains a key research direction.","Uncertainty Quantification in Machine Learning for Biosignal Applications  \nA Review  \nIvo Pascal de Jonga,∗, Andreea Ioana Sburleaa , Matias Valdenegro-Toroa  \na Department of Artificial Intelligence, Bernoulli Institute, University of Groningen, Nijenborg 9, 9747 AG, Groningen, The Netherlands  \nAbstract  \nUncertainty Quantification (UQ) has gained traction in an attempt to improve the interpretability and robustness of machine learning predictions. Specifically (medical) biosignals such as electroencephalography (EEG), electrocardiography (ECG), electrooculography (EOG), and electromyography (EMG) could benefit from good UQ, since these suffer from a poor signal-to-noise ratio, and good human interpretability is pivotal for medical applications. In this paper, wereview the state of the art of applying Uncertainty Quantification to Machine Learning tasks in the biosignal domain. We present various methods, shortcomings, uncertainty measures and theoretical frameworks that currently exist in this application domain. We address misconceptions in the field, provide recommendations for future work, and discuss gapsin the literature in relation to diagnostic implementations as well as control for prostheses or brain-computer interfaces. Overall it can be concluded that promising UQ methods are available, but that research is needed on how people and systems may interact with an uncertainty-model in a (clinical) environment.  \nKeywords: Uncertainty Quantification, Bayesian Neural Networks, Biosignals, EEG, ECG, EOG, EMG, BCI  \n1. Introduction  \nStandard Machine Learning (ML) systems such as Random Forests, SVMs, and Neural Networks typically produce single-point estimates for their classification task. Such single-point models neglect alternative predictions that are consistent with the training data, and therefore give an inadequate estimate of the uncertainty of a prediction. As a result, they may give overconfident but completely inaccurate predictions, which induces skepticism and hinders the implementation of Machine Learning methods in clinical settings [1] . Uncertainty Quantification (UQ) attempts to address this problem by adapting Machine Learning systems to also predict a measure of confidence for a given prediction. Over the past years this has been gaining traction in Computer Vision [2], but it is still only lightly explored in Machine Learning tasks that focus on Biosignals.  \nwork with a Machine Learning model requires UQ to indicate when the model does not know and minimise misclassifications. To give an order of scale to the human effort: sleep scoring a patient’s EEG recording of an overnight stay will typically take a neurologist about two hours [3] . A Machine Learning system that can automatically classify the majority of the overnight stay with high confidence while identifying the parts that it is uncertain on may reduce this.  \nFigure 1 shows various roles uncertainty estimation can play in a biosignal Machine Learning system. The primary use cases are to improve transparency of predictions for a decision support system, or to make independent classifications only when it is likely to be correct. Additionally, uncertainty estimates may be used in various ways to improve the predictions of a Machine Learning model, and it may even be used to determine when ad-  \narXiv :2312 .09454v2 [ ee ss . SP] 4 Jun 2025  \nApplications using biosignals can gain particular benefits from uncertainty quantification. Their signals are sensitive to artifacts that could corrupt the prediction of a Machine Learning system in unexpected ways. Uncertainty Quantification methods may help here by recognizing that the data is corrupted and indicate increased uncertainty.  \nAnother argument for the importance of Uncertainty Quantification is that the human interpretation of the signal requires substantial time investment. Automating this  \n∗ Corresponding author  \nEmail address: [ivo.de.jong@rug.nl](ivo.de.jong@rug.nl) (Ivo Pascal de","cbCaio189FlNg2En","https://ap.wps.com/l/cbCaio189FlNg2En","pdf",1359742,1,30,"English","en",105,"# Introduction\n## Uncertainty in standard machine learning predictions\n## Roles of uncertainty estimation in biosignal systems","[{\"question\":\"Why does uncertainty quantification matter for biosignal machine learning?\",\"answer\":\"Biosignals such as EEG, ECG, EOG, and EMG have poor signal-to-noise ratios and are sensitive to artifacts. UQ helps detect corrupted data and provides a confidence measure that supports medical interpretation.\"},{\"question\":\"What problem do single-point ML models create in this context?\",\"answer\":\"Standard ML models output single-point estimates and ignore alternative consistent predictions. This can produce overconfident yet incorrect results, which limits clinical trust and implementation.\"},{\"question\":\"What roles can uncertainty estimates play in biosignal decision-making?\",\"answer\":\"Uncertainty can enable decision support transparency, support rejection when the model is uncertain, improve ML classification performance, and trigger additional recordings or tests aligned with clinician uncertainty.\"}]","Uncertainty Quantification in Machine Learning for Biosignal Applications - A Review | PDF",1785807684,76,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"uncertainty-quantification-in-machine-learning-for-biosignal-applications-a-review","",{"@graph":36,"@context":86},[37,54,69],{"@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/uncertainty-quantification-in-machine-learning-for-biosignal-applications-a-review/121909/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-04",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 does uncertainty quantification matter for biosignal machine learning?","Question",{"text":76,"@type":77},"Biosignals such as EEG, ECG, EOG, and EMG have poor signal-to-noise ratios and are sensitive to artifacts. UQ helps detect corrupted data and provides a confidence measure that supports medical interpretation.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What problem do single-point ML models create in this context?",{"text":81,"@type":77},"Standard ML models output single-point estimates and ignore alternative consistent predictions. This can produce overconfident yet incorrect results, which limits clinical trust and implementation.",{"name":83,"@type":74,"acceptedAnswer":84},"What roles can uncertainty estimates play in biosignal decision-making?",{"text":85,"@type":77},"Uncertainty can enable decision support transparency, support rejection when the model is uncertain, improve ML classification performance, and trigger additional recordings or tests aligned with clinician uncertainty.","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":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":21,"slug":122},"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":107,"slug":138},19,"General","general"]