Samsung health AI models analyse wearable biosignal data

Samsung Research America’s Digital Health Team has introduced two AI basis models designed to be taught from wearable biosignals. The work centres on data captured by smartwatches, together with coronary heart exercise, sleep, and bodily exercise.

The firm mentioned its Connected Care imaginative and prescient on the Health Forum throughout Galaxy Unpacked in July 2026. Samsung described a way forward for preventive, personalised, and linked care, supported by health know-how and healthcare partnerships. Its analysis staff positions health basis models as one part of latest shopper health experiences.

Sharanya Desai, Head of Digital Health Algorithms at Samsung Research America, mentioned: “This analysis is important as a result of it lays the technical groundwork for delivering health insights which are environment friendly, exact, and steady by way of a health basis mannequin.

“We will proceed to develop and advance health basis models that may be utilized to a wide range of biosignals and health options that may function on-device with restricted sensors and computing assets.”

Samsung’s health AI basis mannequin analysis

A health basis mannequin makes use of self-supervised studying to determine options in unlabeled biosignal data. Samsung says that pretraining on giant health datasets permits one mannequin to assist downstream duties akin to biosignal evaluation, biomarker improvement, and health situation prediction.

The analysis covers two models with completely different goals. xMAE, quick for Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning, learns temporal relationships between completely different biosignals. HiMAE, or Hierarchical Masked Autoencoder, learns health patterns throughout a number of time scales in wearable time-series data.

Samsung says xMAE was accepted to the International Conference on Machine Learning. HiMAE was accepted to the International Conference on Learning Representations. The firm describes each as work on physiological relationships and temporal buildings in biosignal data.

The models tackle completely different elements of wearable-data evaluation. xMAE connects two cardiac indicators that measure associated exercise by way of completely different mechanisms. HiMAE analyses data at quick and lengthy intervals, permitting one pretrained mannequin to assist classification, numerical prediction, and data technology.

xMAE hyperlinks steady PPG data to ECG indicators

Electrocardiograms, or ECGs, measure the center’s electrical exercise instantly. Samsung describes ECG as helpful for measuring coronary heart price and heart-rate variability. It may also determine irregular coronary heart rhythms and dangers related to circumstances akin to atrial fibrillation.

Wearable ECG readings usually require a person to pause and take an energetic measurement. Photoplethysmography, or PPG, takes a special strategy. PPG detects modifications in blood move and may run passively by way of sensors in wearable units akin to smartwatches.

Both indicators originate from cardiac exercise. They happen with a time distinction, which Samsung compares with listening to thunder after seeing lightning. xMAE learns that temporal relationship by reconstructing masked elements of an ECG sign from PPG data.

This design goals to analyse cardiovascular-health options by way of repeatedly measured PPG data with out separate guide ECG measurements. The mannequin’s pretraining used about 9,400 hours of ECG and PPG data.

Subbu Venkatraman, Head of the Digital Health Research Lab at Samsung Research America, commented: “Biosignals are inherently dynamic, with distinctive time-varying physiological properties. The key contribution of this analysis lies in proving the viability of health basis models able to capturing each the inter-signal relationships and their underlying temporal buildings.

“We stay dedicated to advancing foundational health AI analysis and translating it into healthcare options that meaningfully enhance individuals’s health and wellbeing.”

Samsung reviews that xMAE outperformed unimodal biosignal models and present multimodal studying strategies in 15 of 19 analysis duties. Those duties lined heart problems prediction, irregular test-result detection, and sleep-stage classification. The firm additionally says the discovered options confirmed potential to be used throughout sensor units, physique places, and data-gathering environments.

HiMAE analyses wearable data throughout time scales

Wearable data can carry completely different data over completely different time intervals. Short segments can present fast-changing indicators akin to heartbeats. Longer segments can reveal patterns that construct over time, akin to sleep or bodily exercise.

HiMAE makes use of a number of encoders to analyse quick and lengthy data segments individually. Samsung says this association allows the mannequin to determine the time scale wanted for a health process. Heart-rate evaluation and sleep prediction can due to this fact draw on completely different elements of the time-series data.

The coaching technique reconstructs masked parts of wearable data. Samsung says this lets HiMAE be taught patterns from biosignals the place labelled data is restricted. The mannequin then helps classification, numerical prediction, and data technology from a single pretrained system.

Samsung says HiMAE achieved excessive efficiency with a smaller mannequin than present models. The firm additionally reviews that it could possibly produce leads to lower than one millisecond on a smartwatch-class central processing unit.

That processing declare locations the mannequin’s evaluation on the gadget moderately than on cloud servers. Foundation models skilled on unlabelled physiological streams present a mechanism to extract diagnostic markers, run predictive health classifications, and generate person steering from shopper {hardware} all with out steady server connectivity.

See additionally: Google AI health coach to use Abbott glucose data

Want to be taught extra about AI and massive data from business leaders? Check out AI & Big Data Expo going down in Amsterdam, California, and London. The complete occasion is a part of TechEx and is co-located with different main know-how occasions together with the Cyber Security & Cloud Expo. Click here for extra data.

AI News is powered by TechForge Media. Explore different upcoming enterprise know-how occasions and webinars here.

The put up Samsung health AI models analyse wearable biosignal data appeared first on AI News.