[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84564-en":3,"doc-seo-84564-105":29,"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":13,"seo_description":14,"update_tm":27,"read_time":28},84564,8796095360427,"Lucas Martin","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Timesynth: A Temporal Fidelity Framework for Health Signal Digital Twins","Forecasting models for health-signal digital twins must preserve oscillatory timing, frequency, phase, and state-transition dynamics, yet standard pointwise metrics cannot reveal when these properties are lost. Across 11 architectures, similar amplitude or value errors can still produce large phase divergence—up to 53° (≈123 ms at 1.2 Hz)—undetected by conventional benchmarks. TimeSynth introduces a controlled benchmarking framework with a physiologically grounded generator and diagnostics measuring amplitude, frequency, phase, and state-transition fidelity.","arXiv :2607 .0043 1v 1 [ cs .LG] 1 Jul 2026  \nTimesynth: A Temporal Fidelity Framework for Health Signal Digital Twins  \nMd Rakibul Haque 1,2 , Shireen Elhabian 1,2 , Warren Woodrich Pettine3*  \n1 Scientific Computing and Imaging Institute, University of Utah, Salt Lake City, 84112, UT, USA.  \n2 Kahlert School of Computing, University of Utah, Salt Lake City, 84112, UT, USA.  \n3* Department of Psychiatry, University of Utah, Salt Lake City, UT,  \nUSA.  \n*Corresponding author(s). E-mail(s): [warren.pettine@hsc.utah.edu](warren.pettine@hsc.utah.edu) ; Contributing authors: [rakibul.haque@utah.edu](rakibul.haque@utah.edu) ; [shireen@sci.utah.edu](shireen@sci.utah.edu) ;  \nAbstract  \nForecasting models for health-signal digital twins must preserve the oscillatory, frequency, phase, and state-transition dynamics of physiological signals, yet the pointwise metrics used to benchmark them cannot detect when these fundamental properties are lost. We show that this blind spot misranks models: across 11 architectures, models with comparable pointwise error diverge by up to 53◦ in phase accuracy, equivalent to roughly 123 ms for a 1.2 Hz cardiac rhythm and invisible to standard metrics. To enable development of models that escape such failures, we introduce TimeSynth, a controlled benchmarking framework with two reusable components: a physiologically grounded generator producing signals with analytically known ground-truth dynamics from parametric models fitted to real electroencephalography, electrocardiography and photoplethysmogram signals, along with diagnostics quantifying amplitude, frequency, phase, and state-transition fidelity. Linear and full-sequence attention models systematically lose frequency and phase information despite acceptable amplitude error, whereas architectures with localized temporal structure better preserve dynamical fidelity and adapt to observable state transitions; none, however, reliably preserves stochastic switching. Because the dominant determinant of fidelity is architectural, model choice becomes a principled, use-case-driven decision rather  \n1  \nthan a search for a single winner. TimeSynth thus supplies the controlled preclinical stress test missing before models are coupled to patient data, with areusable generator and diagnostics for fidelity-aware development.  \nKeywords: Digital Twin , Forecasting Models, Physiological Signals, State Change  \n1 Introduction  \nForecasting physiological time series underpins applications from continuous patient monitoring and early warning systems to emerging health-signal digital twins [1–3]: patient-specific computational systems, continuously updated with physiological data, that are expected to predict future biological states [4] . Across these applications, forecasting models must do more than predict future values accurately. Physiological signals encode clinically relevant information in their temporal dynamics, including oscillatory structure, frequency content, phase relationships, and transitions between physiological states that evolve across multiple timescales [5–7] . A model that reproduces future values while distorting these dynamics may appear numerically accurate yet fail to preserve the physiological behavior it is meant to represent.  \nYet modern forecasting models are typically evaluated with pointwise metrics such as mean squared error (MSE) and mean absolute error (MAE) [8, 9], which quantify numerical agreement between predicted and observed values but reveal little about whether a forecast preserves oscillatory timing, frequency evolution, or phase coherence [9, 10] . Models with similar forecasting error may therefore represent the underlying process very differently: one may reproduce amplitude while drifting in phase, another may preserve dominant frequencies while smoothing clinically meaningful structure. This is compounded because the dynamical properties of real recordings are generally unknown; without ground-truth amplitude, frequ","cbCaici59l1qmlv3","https://ap.wps.com/l/cbCaici59l1qmlv3","pdf",6537558,1,55,"English","en",105,"# Abstract\n# Keywords\n# Introduction\n## Forecasting for digital twins\n## Limitations of pointwise metrics\n## Need for controlled benchmarking","[{\"question\":\"Why do pointwise metrics like MSE and MAE fail for health-signal digital twin forecasting?\",\"answer\":\"They measure numerical agreement at each time point but do not capture whether forecasts preserve oscillatory timing, frequency evolution, or phase coherence. As a result, models can look accurate while drifting in phase or losing temporal structure.\"},{\"question\":\"What problem does TimeSynth address in evaluating physiological forecasting models?\",\"answer\":\"It targets the gap between forecasting accuracy and physiological fidelity, where models may misranked due to blind spots in standard evaluation. TimeSynth uses controlled signals with known dynamics and fidelity diagnostics to expose these failures.\"},{\"question\":\"How does TimeSynth assess fidelity beyond standard error?\",\"answer\":\"It combines a generator that produces signals with analytically known ground-truth dynamics from parametric fits to real EEG, ECG, and PPG data, with diagnostics that quantify amplitude, frequency, phase, and state-transition fidelity.\"}]",1784196840,139,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"timesynth-a-temporal-fidelity-framework-for-health-signal-digital-twins","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/timesynth-a-temporal-fidelity-framework-for-health-signal-digital-twins/84564/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-22","2026-07-16",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 do pointwise metrics like MSE and MAE fail for health-signal digital twin forecasting?","Question",{"text":75,"@type":76},"They measure numerical agreement at each time point but do not capture whether forecasts preserve oscillatory timing, frequency evolution, or phase coherence. As a result, models can look accurate while drifting in phase or losing temporal structure.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What problem does TimeSynth address in evaluating physiological forecasting models?",{"text":80,"@type":76},"It targets the gap between forecasting accuracy and physiological fidelity, where models may misranked due to blind spots in standard evaluation. TimeSynth uses controlled signals with known dynamics and fidelity diagnostics to expose these failures.",{"name":82,"@type":73,"acceptedAnswer":83},"How does TimeSynth assess fidelity beyond standard error?",{"text":84,"@type":76},"It combines a generator that produces signals with analytically known ground-truth dynamics from parametric fits to real EEG, ECG, and PPG data, with diagnostics that quantify amplitude, frequency, phase, and state-transition fidelity.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"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":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]