Federated Learning (FL) is a new technology that has been a hot research topic. It enables training an algorithm across multiple decentralized edge devices or servers holding local data samples, without exchanging them. There are many application domains where large amounts of properly labeled and complete data are not available in a centralized location, for example, doctors' diagnosis from medical image analysis. There are also growing concerns over data and user privacy as Artificial Intelligence is becoming ubiquitous in new application domains. As such, very recently, a lot of research has been conducted in several areas within the nascent field of FL. A variety of surveys on different subtopics exist in current literature, focusing on specific challenges, design aspects and application domains. In this paper, we review existing contemporary works in the related areas in order to understand the challenges and topics that are emphasized by each type of FL surveys. Furthermore, we categorize FL research in terms of challenges, design factors and applications, conducting a holistic review of each and outlining promising research directions.
Objective:To examine whether a higher number of virtual interactions with health coaches through a digital health platform would lead to a greater reduction in blood pressure than people with a lower number of interactions.Design and method:The participants were 447 people with hypertension or a baseline blood pressure value of 140/90 mmHg or higher (female, 77.6%. male, 20.8%. Average age 43.9 years) enrolled on the CompleteHealthô platform between April 2018 and April 2021 who had entered a subsequent blood pressure value after 6 months of enrollment. The change in blood pressure was calculated as the difference between their blood pressure after 6 months and their first blood pressure value entered on CompleteHealthô. Engagement after 6 months was defined across two levels: minimally engaged members who had at most 1 interaction with a health coach and highly engaged members who had more than 1 interaction with a health coach. Given that the members are generally low-income people, we determined more than 1 interaction as a significant threshold for engagement.By enrolling on mDoc Healthcare’s digital platform, CompleteHealthô, people were given access to self-care support via a virtual coach. They tracked their health metrics via the app, were given access to tele-education classes and community forums, and the coach provided health education and digital nudges on healthy behaviours to take, including regular blood pressure checks.Results:After 6 months, of the 447 people, 374 (83.7%) were highly engaged while 73 (16.3%) were minimally engaged. Highly engaged members saw an average reduction in their systolic and diastolic blood pressures of 13.6 mmHg and 10.4 mmHg respectively, while minimally engaged members saw reductions of 5.8 mmHg and 4.9 mmHg in their systolic and diastolic blood pressures. Limitations were that this did not take into account asynchronous engagement on the platform e.g. self-tracking, review of education messages etc.Conclusions:The results of the retrospective study show that increased engagement with a digital self-care intervention drives a higher reduction in blood pressure for people living with hypertension or an elevated blood pressure level.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.