[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122480-en":3,"doc-seo-122480-105":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},122480,1374391974585,"Genevieve","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Wearable edge machine learning with synthetic photoplethysmograms - Research overview","Wearable health technologies face strict privacy requirements alongside biased or insufficient training datasets, making robust machine learning hard to develop and deploy. Parametric synthetic data can reproduce realistic signals with controllable information content, while avoiding direct exposure of sensitive data from the observable world. This article presents a system combining a synthetic photoplethysmogram generator, CNN models trained on synthetic signals, an edge device estimating heart rate from real-time PPG, and a mobile app, achieving accurate foot detection with privacy preserved.","Expert Systems With Applications 238 (2024) 121523  \n| Wearable edge machine learning with synthetic photoplethysmograms✩ Jukka-Pekka Sirkiä ∗, Tuukka Panula, Matti Kaisti\u003Cbr>Department of Computing, University of Turku, Vesilinnantie 5, Turku, 20500, Finland |  |\n| --- | --- |\n| A R T I C L E I N F O | A B S T R A C T\u003Cbr>Strict privacy regulations pose challenges to the development of machine learning (ML) in the field of health technology where data is particularly sensitive. Gathering and using robust, bias-free, and suitably anonymized datasets required by ML models is difficult, time-consuming, and thus expensive. Parametric synthetic data offers a solution by mimicking real-world processes with easily adjustable parameters that shape the information content of the data as desired. This article presents a system demonstrating how synthetic data can be used in conjunction with wearable edge devices. Importantly, the system preserves privacy as thereis no risk of leaking sensitive information from the model or during the use of the wearable device. The system consists of (1) a synthetic photoplethysmogram (PPG) model,(2) convolutional neural network (CNN) models trained with the synthetic signals,(3) a wearable edge device that computes heart rate from real-time PPG signals using the developed CNN models, and (4) an accompanying mobile phone application receiving the results. The synthetic model produces realistic PPG signals together with labels that can be used in CNN model training. The quality of the synthetic data is sufficient to train even a tiny CNN model with only two convolutional layers and 28 parameters to detect PPG waveform feet. The developed wearable device is able to run the model smoothly and the performance of the model is on par with the more complex models and other foot detection algorithms. |\n| Dataset link: [https://github.com/jpsirk/synthet](https://github.com/jpsirk/synthet)[ic-photoplethysmograms](ic-photoplethysmograms) |  |\n| Keywords:\u003Cbr>Edge machine learning Neural network Wearable Photoplethysmography Synthetic |  |\n\n1. Introduction  \nThe use of machine learning (ML) methods in the field of health technology is typically hindered by privacy requirements, biased training data and insufficient amount of training data (Chen et al., 2021; Rajotte et al., 2022). Tightened privacy regulations, such as the General Data Protection Regulation (GDPR) 2016/679 of the European Union (EU), have led some to demand to remove the barriers to sharing health data to avoid potentially damaging effects on research and therefore ultimately on patients (Bentzen et al., 2021). For example, pseudonymized data are considered personal data under GDPR eventhough the identifiers have been replaced by codes (Bentzen et al., 2021). Bias in medical data can easily occur if the study participant backgrounds are homogeneous (e.g., socioeconomic factors, gender, age, and diseases) or the classes present in the data are imbalanced. Gathering data with only one type of equipment is also prone to producing data that might not generalize well in reality Chen et al. (2021). The lack of suitable existing data or access to them can also pose difficulties in obtaining large datasets. Data sharing practices are intended to facilitate the verification of published results and help build on existing data, but challenges arise from obeying these practices (Naudet et al., 2018).  \nThese challenges can be addressed with synthetic data, which is data not originated directly from the observable world. Synthetic data can be generated with ML-based generative models (e.g. generative adversarial networks (GANs)), parametric models or physical simulations (Chenet al., 2021; Jordon et al., 2022). The models-based approach allows to efficiently generate data in large quantities, and the problem of bias is easier to tackle. More importantly, parametric and physical simulation models are inherently private, given that the data is generated with pure","cbCaiiuFdlk9YqZ8","https://ap.wps.com/l/cbCaiiuFdlk9YqZ8","pdf",1225524,1,10,"English","en",105,"# Introduction\n## Privacy and data quality challenges in health ML\n## Synthetic data as a privacy-preserving solution\n## Edge ML enablement\n# System architecture with synthetic PPG and wearable edge inference\n## Synthetic PPG generation and labeling\n## CNN training with synthetic signals\n## Wearable edge device inference\n## Mobile application delivery","[{\"question\":\"Why are privacy regulations a barrier for machine learning in health technology?\",\"answer\":\"Health ML development is hindered by sensitivity of data and requirements such as GDPR. Sharing and using robust datasets becomes difficult, time-consuming, and costly under these rules.\"},{\"question\":\"How does the approach use synthetic photoplethysmograms to address dataset problems?\",\"answer\":\"Synthetic data is generated to mimic real-world processes with adjustable parameters. This enables creating large labeled datasets while simplifying bias mitigation compared with relying on limited real-world data.\"},{\"question\":\"What components make up the wearable edge system described in the article?\",\"answer\":\"The system includes a synthetic PPG model, CNN models trained on synthetic signals, a wearable edge device that computes heart rate from real-time PPG using the CNN, and a mobile phone application that receives the results.\"}]","Wearable edge machine learning with synthetic photoplethysmograms - Research overview | PDF",1785810874,25,{"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},"wearable-edge-machine-learning-with-synthetic-photoplethysmograms-research-overview","",{"@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/wearable-edge-machine-learning-with-synthetic-photoplethysmograms-research-overview/122480/",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","2026-08-04",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 are privacy regulations a barrier for machine learning in health technology?","Question",{"text":75,"@type":76},"Health ML development is hindered by sensitivity of data and requirements such as GDPR. Sharing and using robust datasets becomes difficult, time-consuming, and costly under these rules.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the approach use synthetic photoplethysmograms to address dataset problems?",{"text":80,"@type":76},"Synthetic data is generated to mimic real-world processes with adjustable parameters. This enables creating large labeled datasets while simplifying bias mitigation compared with relying on limited real-world data.",{"name":82,"@type":73,"acceptedAnswer":83},"What components make up the wearable edge system described in the article?",{"text":84,"@type":76},"The system includes a synthetic PPG model, CNN models trained on synthetic signals, a wearable edge device that computes heart rate from real-time PPG using the CNN, and a mobile phone application that receives the results.","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},[93,97,101,105,110,115,120,123,128,131,134],{"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":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":21,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":21,"slug":133},"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]