Litcius/Paper detail

Self-supervised transfer learning of physiological representations from free-living wearable data

Dimitris Spathis, Ignacio Perez-Pozuelo, Søren Brage, Nicholas J. Wareham, Cecilia Mascolo

202140 citationsDOIOpen Access PDF

Abstract

Wearable devices such as smartwatches are becoming increasingly popular tools for objectively monitoring physical activity in free-living conditions. To date, research has primarily focused on the purely supervised task of human activity recognition, demonstrating limited success in inferring high-level health outcomes from low-level signals. Here, we present a novel self-supervised representation learning method using activity and heart rate (HR) signals without semantic labels. With a deep neural network, we set HR responses as the supervisory signal for the activity data, leveraging their underlying physiological relationship. In addition, we propose a custom quantile loss function that accounts for the long-tailed HR distribution present in the general population.

Topics & Concepts

Computer scienceWearable computerSmartwatchLeverage (statistics)Machine learningArtificial intelligenceActivity recognitionTransfer of learningWearable technologyMulti-task learningTask (project management)Deep learningAccelerometerFeature learningSupervised learningArtificial neural networkEmbedded systemOperating systemEconomicsManagementContext-Aware Activity Recognition SystemsNon-Invasive Vital Sign MonitoringPhysical Activity and Health
Self-supervised transfer learning of physiological representations from free-living wearable data | Litcius