Context-aware support for cardiac health monitoring using federated machine learning

Context-awareness provides a platform for healthcare professionals to assess the health status of patients in their care using multiple relevant parameters such as heart rate, electrocardiogram (ECG) signals and activity data. It involves the use of digital technologies to monitor the health condition of a patient in an intelligent environment. Feedback gathered from relevant professionals at earlier stages of the project indicates that physical activity recognition is an essential part of cardiac condition monitoring. However, the traditional machine learning method f developing a model for activity recognition suffers two significant challenges; model overfitting and privacy infringements. This research proposes an intelligent and privacy-oriented context-aware decision support system for cardiac health monitoring using the physiological and the activity data of the patient. The system makes use of a federated machine learning approach to develop a model for physical activity recognition. Experimental analysis shows that the federated approach has advantages over the centralized approach in terms of model generalization whilst maintaining the privacy of the user.

This version of the contribution has been accepted for publication, after peer review (when applicable) but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/10.1007/978-3-030-91100-3_22 Use of this Accepted Version is subject to the publisher’s Accepted Manuscript terms of use https://www.springernature.com/gp/open-research/policies/accepted-ma...