When we talk to clinicians and psychologists, it becomes evident that developing even relatively simple mHealth apps, let alone state-of-the-art EMA and EMI apps that aim to systematically and unobtrusively collect sensor and/or patient data, analyse it, and act on it in (near) real-time, is a challenging, expensive, and time-consuming endeavour for the mental health clinical practice and research communities.

A few months ago, GEOTEC members were glad to report on the NTD (NativeScript Task Dispatcher) library for systematic data collection using mobile devices, which circumvented the restrictions of mobile operating systems and manufacturers, to ensure reliable and sustained passive detection over time. NTD was a first but important step towards the next generation of mHealth applications that GEOTEC envisions as part of the series of research projects under the umbrella research iniciative called SyMptOMS.

Today, we are proud to achieve a significant step to pave the way for unfolding the next-generation EMA and EMI apps for mental health. Led by Alberto as part of his PhD project, a group of GEOTEC researchers have developed the AwarNS Framework, an open source, modular, reusable and extensible context-aware development framework based on Android.

AwarNS supersedes the NTD library to provide the necessary data collection mechanisms, data representation models, real-time analysis, and notification support to create advanced mobile applications for mental health monitoring and interventions. In addition, it works only on the phone, yes, with no external server-side dependencies, making it suitable for offline and strict privacy scenarios. By default, sensed data is only stored on the mobile device, unless the user/patient optionally can synchronise it with a remote server.

Therefore, AwarNS is especially targeted at mHealth application developers, who design and build applications to monitor and detect changes in a wide variety of contextual sources offered by smartphones. AwarNS hides the technical complexity of the mobile operating systems and underlying sensor APIs, and considerably speeds up the development process for this kind of apps.

In a recent open access paper in the Journal of Biomedical Informatics , we describe the AwarNS’ design principles and architecture design, explain its capabilities and implementation, and demonstrate its use in three real-life case studies implementing various mobile interventions for different mental disorders used in clinical practice. Rather than reproducing the abstract here, the items below, extracted from the paper, seem much more significant to grasp the novelty of the paper.

• Problem: There exists no generic solution to reliably sample patients’ context through smartphones, analyse the data and react to its changes on the device, which is critical for the success of digital phenotyping and just-in-time assessments and interventions.
• What’s already known: Existing tools can sample the context, but they are either intrusive or struggle with phone OS’ background execution restrictions. Only a few can deliver assessments and interventions but rely on external servers to analyse the data.
• What AwarNS adds: A versatile, reusable framework to develop reactive mHealth applications, based on reliable systematic context sampling while preserving privacy, to enable timely (offline) assessments and interventions.

Full citation:

Alberto González-Pérez, Miguel Matey-Sanz, Carlos Granell, Laura Díaz-Sanahuja, Juana Bretón-López, Sven Casteleyn (2023) AwarNS: A framework for developing context-aware reactive mobile applications for health and mental health. Journal of Biomedical Informatics, 141, 104359, doi: 10.1016/j.jbi.2023.104359.