JMIR Publications

ABSTRACT

Background:

The potential of digital mental health to transform care delivery in low and middle-income countries is well established. However, one barrier is the need to clinically and culturally adapt current tools to local needs. This article explores one example of this process by creating a novel data visualization system for a digital mental health app and outlines the necessary steps in the process.

Objective:

Our objective is to develop tools to make inferences from the data obtained from digital phenotyping for schizophrenia. The tool is specific in its scope of being extendible from an already existing mindLAMP platform, and it allows exploratory analysis for sparse data from digital phenotyping. While the data is voluminous when collected passively from sensors on the mobile phone, the number of engaged users providing active data is sparse. Particularly, the data on relapses has uncertainties. Altogether, the proposed tool should appropriately handle data, and provide exploratory visualizations to compare the behavior of a patient across different modalities or time instances, and also trends across different patients. Also, the tool is designed to accommodate the requirements put forth by the teams in Bangalore and Boston involved in this work.

Methods:

We adapted the mindLAMP app, already used in many countries today, to offer a new data visualization portal, mindLAMPVis. The portal is designed to improve clinical integration for use in India. After building the new portal, we corroborate insights from this new portal with known clinical observations of relapse, to support the use of comparative visualization. We then worked to visualize these insights by taking data from the mindLAMP app and using multivariate analysis and dimensionality reduction to make the data easy for clinicians and patients to interact with. We integrate these techniques in a novel interactive visualization tool, mindLAMPVis, a locally co-designed, developed, and deployed tool.

Results:

To assess the system, we preloaded data of 24 patients with schizophrenia, including those with relapses. Through case examples focusing on relapse risk prediction in schizophrenia, we utilized mindLAMP vis to identify different visualization methods to compare different analytical results for each patient. In partnership with clinicians, we then explored the clinical potential of mindLAMPvis to inform care. As an example of reverse translation, we found that mindLAMPvis offers new features that complement to original features in mindLAMP – highlighting the mutual benefit of software adaptation and co-design.

Conclusions:

mindLAMPVis is a tailored tool designed for use in India, but it can aid in identifying and comparing behavioral patterns that may indicate clinical risk for patients in any country. mindLAMPVis offers an example of how through technical, design, feedback, and real-world clinical testing it is feasible to adapt current software tools to meet local needs and even exceed the use cases of the original technology. Clinical Trial: Not applicable