End-to-end Personalization of Digital Health Interventions using Raw Sensor Data with Deep Reinforcement Learning

2019 ◽  
Author(s):  
Ali el Hassouni ◽  
Mark Hoogendoorn ◽  
A.E. Eiben ◽  
Martijn van Otterlo ◽  
Vesa Muhonen
Keyword(s):  
Reinforcement Learning ◽  
Digital Health ◽  
Health Interventions ◽  
Sensor Data ◽  
End To End

scholarly journals Verifiably safe exploration for end-to-end reinforcement learning

2021 ◽  
Author(s):  
Nathan Hunt ◽  
Nathan Fulton ◽  
Sara Magliacane ◽  
Trong Nghia Hoang ◽  
Subhro Das ◽  
...  
Keyword(s):  
Reinforcement Learning ◽  
End To End

scholarly journals The role of economic, educational and social resources in supporting the use of digital health technologies by people with T2D: a qualitative study

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Sophie Turnbull ◽  
Patricia J. Lucas ◽  
Alastair D. Hay ◽  
Christie Cabral
Keyword(s):  
Social Networks ◽  
Social Status ◽  
Digital Health ◽  
Self Care ◽  
Social Resources ◽  
Health Interventions ◽  
Structured Interviews ◽  
Health Technologies ◽  
Digital Skills ◽  
Care Support

Abstract Background Type 2 Diabetes (T2D) is a common chronic disease, with socially patterned incidence and severity. Digital self-care interventions have the potential to reduce health disparities, by providing personalised low-cost reusable resources that can increase access to health interventions. However, if under-served groups are unable to access or use digital technologies, Digital Health Technologies (DHTs) might make no difference, or worse, exacerbate health inequity. Study aims To gain insights into how and why people with T2D access and use DHTs and how experiences vary between individuals and social groups. Methods A purposive sample of people with experience of using a DHT to help them self-care for T2D were recruited through diabetes and community groups. Semi-structured interviews were conducted in person and over the phone. Data were analysed thematically. Results A diverse sample of 21 participants were interviewed. Health care practitioners were not viewed as a good source of information about DHTs that could support T2D. Instead participants relied on their digital skills and social networks to learn about what DHTs are available and helpful. The main barriers to accessing and using DHT described by the participants were availability of DHTs from the NHS, cost and technical proficiency. However, some participants described how they were able to draw on social resources such as their social networks and social status to overcome these barriers. Participants were motivated to use DHTs because they provided self-care support, a feeling of control over T2D, and personalised advice or feedback. The selection of technology was also guided by participants’ preferences and what they valued in relation to DHTs and self-care support, and these in turn were influenced by age and gender. Conclusion This research indicates that low levels of digital skills and high cost of digital health interventions can create barriers to the access and use of DHTs to support the self-care of T2D. However, social networks and social status can be leveraged to overcome some of these challenges. If digital interventions are to decrease rather than exacerbate health inequalities, these barriers and facilitators to access and use must be considered when DHTs are developed and implemented.

scholarly journals Digital health interventions in palliative care: a systematic meta-review

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Anne M. Finucane ◽  
Hannah O’Donnell ◽  
Jean Lugton ◽  
Tilly Gibson-Watt ◽  
Connie Swenson ◽  
...  
Keyword(s):  
Decision Making ◽  
Palliative Care ◽  
Systematic Reviews ◽  
Psychological Symptoms ◽  
Digital Health ◽  
Evidence Synthesis ◽  
Information Provision ◽  
Health Interventions ◽  
Meta Review

AbstractDigital health interventions (DHIs) have the potential to improve the accessibility and effectiveness of palliative care but heterogeneity amongst existing systematic reviews presents a challenge for evidence synthesis. This meta-review applied a structured search of ten databases from 2006 to 2020, revealing 21 relevant systematic reviews, encompassing 332 publications. Interventions delivered via videoconferencing (17%), electronic healthcare records (16%) and phone (13%) were most frequently described in studies within reviews. DHIs were typically used in palliative care for education (20%), symptom management (15%), decision-making (13%), information provision or management (13%) and communication (9%). Across all reviews, mostly positive impacts were reported on education, information sharing, decision-making, communication and costs. Impacts on quality of life and physical and psychological symptoms were inconclusive. Applying AMSTAR 2 criteria, most reviews were judged as low quality as they lacked a protocol or did not consider risk of bias, so findings need to be interpreted with caution.

Effectiveness of Digital Health Interventions on Unintentional Injury, Violence, and Suicide: Meta-Analysis

2020 ◽  
pp. 152483802096734
Author(s):  
Mengtong Chen ◽  
Ko Ling Chan
Keyword(s):  
Interpersonal Violence ◽  
Care Delivery ◽  
Digital Health ◽  
Meta Analysis ◽  
Random Effect ◽  
Unintentional Injury ◽  
Health Interventions ◽  
Effect Sizes ◽  
Future Research ◽  
Face To Face

Digital technologies are increasingly used in health-care delivery and are being introduced into work to prevent unintentional injury, violence, and suicide to reduce mortality. To understand the potential of digital health interventions (DHIs) to prevent and reduce these problems, we conduct a meta-analysis and provide an overview of their effectiveness and characteristics related to the effects. We searched electronic databases and reference lists of relevant reviews to identify randomized controlled trials (RCTs) published in or before March 2020 evaluating DHIs on injury, violence, or suicide reduction. Based on the 34 RCT studies included in the meta-analysis, the overall random effect size was 0.21, and the effect sizes for reducing suicidal ideation, interpersonal violence, and unintentional injury were 0.17, 0.24, and 0.31, respectively, which can be regarded as comparable to the effect sizes of traditional face-to-face interventions. However, there was considerable heterogeneity between the studies. In conclusion, DHIs have great potential to reduce unintentional injury, violence, and suicide. Future research should explore DHIs’ successful components to facilitate future implementation and wider access.

Sign in / Sign up

Export Citation Format

Share Document