Combining Mobile Crowdsensing and Ecological Momentary Assessments in the Healthcare Domain

Abstract

The increasing prevalence of smart mobile devices (e.g., smartphones) enables the combined use of mobile crowdsensing (MCS) and ecological momentary assessments (EMA) in the healthcare domain. By correlating qualitative longitudinal and ecologically valid EMA assessment data sets with sensor measurements in mobile apps, new valuable insights about patients (e.g., humans who suffer from chronic diseases) can be gained. However, there are numerous conceptual, architectural and technical, as well as legal challenges when implementing a respective software solution. Therefore, the work at hand (1) identifies these challenges, (2) derives respective recommendations, and (3) proposes a reference architecture for a MCS-EMA-platform addressing the defined recommendations. The required insights to propose the reference architecture were gained in several large-scale mHealth crowdsensing studies running for many years and different healthcare questions. To mention only two examples, we are running crowdsensing studies on questions for the tinnitus chronic disorder or psychological stress. We consider the proposed reference architecture and the identified challenges and recommendations as a contribution in two respects. First, they enable other researchers to align our practical studies with a baseline setting that can satisfy the variously revealed insights. Second, they are a proper basis to better compare data that was gathered using MCS and EMA. In addition, the combined use of MCS and EMA increasingly requires suitable architectures and associated digital solutions for the healthcare domain.

Predicting the Health Condition of mHealth App Users with Large Differences in the Number of Recorded Observations - Where to Learn from?

“…As part of two pilot studies on empowerment of diabetes patients, a mobile crowdsensing framework was adjusted to implement the TrackYourDiabetes mHealth platform [4,9]. Figure 1 summarizes the entire procedure of the app from the patient's point of view.…”

Mobile Health App Database - A Repository for Quality Ratings of mHealth Apps

“…Mobile apps and their data collection capabilities are only one direction that has garnered a lot of attention in the last years [1]. Especially in healthcare and medicine, mobile apps are a basis for new data sources and medical insights [2]- [4]. In addition to the sophisticated collection possibilities of data, mobile apps can be utilized to guide and inform patients about health conditions and questions as well as health-related day-by-day issues [5], [6].…”

Efficient Processing of Geospatial mHealth Data Using a Scalable Crowdsensing Platform

“…With mobile crowdsensing, a person’s health status can be efficiently monitored, but beyond this opportunity, each crowd user benefits from the measurements of other crowd users as well since measured values can be compared among the users over time. Especially for users that suffer from a chronic disease or disorder, the approach of mobile crowdsensing can be very helpful [ 1 , 2 , 3 ]. For the tinnitus chronic disorder, the measurement of noise exposure can help each person in a crowd to avoid harmful locations based on the individual measurements and their aggregations.…”

“…Based on this information, places with an unhealthy noise exposure can be, for example, visually highlighted on a map on the smartphone of a user. On the one hand, as potentially millions of noise measurements have to be collected and processed continuously and concurrently, such a backend in the mHealth context has to ensure a high degree of scalability to avoid a degradation of the service for its users [ 3 ]. In addition, since the workload of such a system can change frequently (e.g., due to day times or public events), such a backend should also provide elasticity.…”

“…Moreover, recent works show [ 3 , 4 ] that approaches that combine contemporary mobile technology with paradigms from the healthcare side should address the question how this combination can be accomplished so that various diseases and their particular aspects can be flexibly addressed. To this end, based on the architecture that is presented in this work, including the promising experimental results, the approach may constitute as a solid basis for other disorders than tinnitus.…”