Moushumi Sharmin

Are we there yet?: feasibility of continuous stress assessment via wireless physiological sensors

Stress can lead to headaches and fatigue, precipitate addictive behaviors (e.g., smoking, alcohol and drug use), and lead to cardiovascular diseases and cancer. Continuous assessment of stress from sensors can be used for timely delivery of a variety of interventions to reduce or avoid stress. We investigate the feasibility of continuous stress measurement via two field studies using wireless physiological sensors --- a four-week study with illicit drug users (n = 40), and a one-week study with daily smokers and social drinkers (n = 30). We find that 11+ hours/day of usable data can be obtained in a 4-week study. Significant learning effect is observed after the first week and data yield is seen to be increasing over time even in the fourth week. We propose a framework to analyze sensor data yield and find that losses in wireless channel is negligible; the main hurdle in further improving data yield is the attachment constraint. We show the feasibility of measuring stress minutes preceding events of interest and observe the sensor-derived stress to be rising prior to self-reported stress and smoking events.

Center of excellence for mobile sensor data-to-knowledge (MD2K)

The Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K) is enabling the collection of high-frequency mobile sensor data for the development and validation of novel multisensory biomarkers and sensor-triggered interventions.

Finding Significant Stress Episodes in a Discontinuous Time Series of Rapidly Varying Mobile Sensor Data

Management of daily stress can be greatly improved by delivering sensor-triggered just-in-time interventions (JITIs) on mobile devices. The success of such JITIs critically depends on being able to mine the time series of noisy sensor data to find the most opportune moments. In this paper, we propose a time series pattern mining method to detect significant stress episodes in a time series of discontinuous and rapidly varying stress data. We apply our model to 4 weeks of physiological, GPS, and activity data collected from 38 users in their natural environment to discover patterns of stress in real life. We find that the duration of a prior stress episode predicts the duration of the next stress episode and stress in mornings and evenings is lower than during the day. We then analyze the relationship between stress and objectively rated disorder in the surrounding neighborhood and develop a model to predict stressful episodes.

Visualization of time-series sensor data to inform the design of just-in-time adaptive stress interventions

We investigate needs, challenges, and opportunities in visualizing time-series sensor data on stress to inform the design of just-in-time adaptive interventions (JITAIs). We identify seven key challenges: massive volume and variety of data, complexity in identifying stressors, scalability of space, multifaceted relationship between stress and time, a need for representation at multiple granularities, inter-person variability, and limited understanding of JITAI design requirements due to its novelty. We propose four new visualizations based on one million minutes of sensor data (n=70). We evaluate our visualizations with stress researchers (n=6) to gain first insights into its usability and usefulness in JITAI design. Our results indicate that spatio-temporal visualizations help identify and explain between- and within-person variability in stress patterns and contextual visualizations enable decisions regarding the timing, content, and modality of intervention. Interestingly, a granular representation is considered informative but noise-prone; an abstract representation is the preferred starting point for designing JITAIs.

TimeStitch: Interactive multi-focus cohort discovery and comparison

Whereas event-based timelines for healthcare enable users to visualize the chronology of events surrounding events of interest, they are often not designed to aid the discovery, construction, or comparison of associated cohorts. We present TimeStitch, a system that helps health researchers discover and understand events that may cause abstinent smokers to lapse. TimeStitch extracts common sequences of events performed by abstinent smokers from large amounts of mobile health sensor data, and offers a suite of interactive and visualization techniques to enable cohort discovery, construction, and comparison, using extracted sequences as interactive elements. We are extending TimeStitch to support more complex health conditions with high mortality risk, such as reducing hospital readmission in congestive heart failure.

Understanding the Feasibility of a Location-Aware Mobile-Based 911-Like Emergency Service in Bangladesh

In this paper, we present the design and implementation of a location-aware mobile-based emergency service for Bangladesh, a developing country, which lacks any central 911-like emergency service. Our goal was to investigate the feasibility and acceptability of such services that do not require building any new infrastructure or changing the existing infrastructure available in Bangladesh. To achieve our goal, we iteratively designed and deployed two location-aware mobile-based emergency services in Dhaka, the capital city of Bangladesh. These deployments provided a deep insight about user experience, acceptance, and sustainability issues. We learned that users considered these services effective, felt comfortable using these during emergencies, and expressed a need for integration of additional services. Our findings indicate that it is feasible to design a location-aware mobile-based emergency service in Bangladesh. We believe that our proposed design will help to provide a low-cost alternative to the central 911-services, especially for the developing countries.

Exploratory Visual Analytics of Mobile Health Data: Sensemaking Challenges and Opportunities

With every advancement in mHealth sensing technology, we are presented with an abundance of data streams and models that enable us to make sense of health information we record. To distill this diverse and ever-growing data into meaningful information, we must first develop tools that can represent data intuitively and are flexible enough to handle the special characteristics of mHealth records. For example, whereas traditional health data such as electronic health records (EHR) often consist of discrete events that may be readily analyzed and visualized, mHealth entails sensor ensembles that generate continuous, multivariate data streams of high-resolution and often noisy measurements. Drawing from methodologies in machine learning and visualization, interactive visual analytics tools are an increasingly important aid to making sense of this complexity. Still, these computational and visual techniques must be employed attentively to represent this data not only intuitively, but also accurately, transparently, and in a way that is driven by user needs. Acknowledging these challenges, we review existing visual analytic tools to identify design solutions that are both useful for and adaptable to the demands of mHealth data analysis tasks. In doing so, we identify open problems for representing and understanding mHealth data, suggesting future research directions for developers in the field.

From Research to Practice: Informing the Design of Autism Support Smart Technology

Smart technologies (wearable and mobile devices) show tremendous potential in the detection, diagnosis, and management of Autism Spectrum Disorder (ASD) by enabling continuous real-time data collection, identifying effective treatment strategies, and supporting intervention design and delivery. Though promising, effective utilization of smart technology in aiding ASD is still limited. We propose a set of implications to guide the design of ASD-support technology by analyzing 149 peer-reviewed articles focused on children with autism from ACM Digital Library, IEEE Xplore, and PubMed. Our analysis reveals that technology should facilitate real-time detection and identification of points-of-interest, adapt its behavior driven by the real-time affective state of the user, utilize familiar and unfamiliar features depending on user-context, and aid in revealing even minuscule progress made by children with autism. Our findings indicate that such technology should strive to blend-in with everyday objects. Moreover, gradual exposure and desensitization may facilitate successful adaptation of novel technology.

Bangladesh Emergency Services: A Mobile Application to Provide 911-Like Service in Bangladesh

We present the design, implementation, and evaluation of Bangladesh Emergency Services (BES), a mobile application that provides 911-like services in Bangladesh. The involvement of A2I, a government agency, helped in the nationwide deployment of BES. To date, 120,382 unique users downloaded BES and 27,117 users are actively using it. Analysis of user ratings (N=3,788) and reviews (N=1,231) from Google Play Store, an online survey (N=137), and face-to-face interviews (N=10) revealed that users considered BES effective. Currently, BES is used to report aggressive behavior, street fight, harassment, fire-related incidents, and for locating nearby hospitals for medical emergencies. Our findings highlight the importance of trustworthiness, scalability of service, and iterative and participatory design for successful deployment of such services. We believe our research will inspire the design of mobile-based emergency applications in other developing countries that lack 911-like services.

Chronodes: Interactive Multifocus Exploration of Event Sequences

The advent of mobile health (mHealth) technologies challenges the capabilities of current visualizations, interactive tools, and algorithms. We present Chronodes, an interactive system that unifies data mining and human-centric visualization techniques to support explorative analysis of longitudinal mHealth data. Chronodes extracts and visualizes frequent event sequences that reveal chronological patterns across multiple participant timelines of mHealth data. It then combines novel interaction and visualization techniques to enable multifocus event sequence analysis, which allows health researchers to interactively define, explore, and compare groups of participant behaviors using event sequence combinations. Through summarizing insights gained from a pilot study with 20 behavioral and biomedical health experts, we discuss Chronodess efficacy and potential impact in the mHealth domain. Ultimately, we outline important open challenges in mHealth, and offer recommendations and design guidelines for future research.