PROJECT SUMMARY/ABSTRACT: In Bangladesh, social-cultural-financial constraints and a scarcity of mental health care practitioners has deprived families raising children with Autism Spectrum Disorder (ASD) from regular monitoring, care, and support. However, the overwhelming adoption (~80%) of mobile phones in Bangladesh in recent years has created an opportunity to improve the existing practice of care using affordable mobile applications. We propose to design, build, deploy, and study the impact of mCARE, a mobile application that will introduce Experience Sampling Method (ESM) (an established and effective method of monitoring the progress of mental health) to the existing practice by allowing the caregivers to routinely report the behavioral progress of children with ASD, and work as an evidence-based and data-driven remote behavioral monitoring platform for the care practitioners. The design of mCARE will follow Value Sensitive Design (VSD) approach to ensure a tight integration with the local economic, social, and cultural values in Bangladesh. The objectives of this proposal are as follows: a) design and build mCARE, that will allow caregivers to routinely report, and thus build the personal records of behavioral progress for each child with ASD, b) improve and expedite the decision making process of the physicians and therapists (referred as only practitioners hereafter) by building appropriate visualization tools to summarize this information, and c) assess the impact of mCARE on treatment and management practices around ASD care in Bangladesh. We propose mCARE to be tested with caregivers of 300 children with ASD in Bangladesh for one year. The children will be selected from four reputed ASD care institutions in Bangladesh – two government and two privately owned, situated in two different districts. They are – the National Institute of Mental Health (NIMH), the Institute of Pediatric Neuro-disorder & Autism (IPNA), Autism Welfare Foundation (AWF), and “Nishpap”. The participants will cover a wide range of socio-economic and cultural diversity in Bangladesh. A positive outcome of using mCARE will promote scaling up this service to all of Bangladesh. This study will introduce a novel digital platform for behavioral research for both public and private mental health institutions in Bangladesh. Furthermore, the administrative (public-private collaboration) and technical (mCARE tools) infrastructures developed through this project can later be utilized and/or replicated for long-term monitoring of other chronic diseases including diabetes and AIDS both in Bangladesh and other LMICs around the world. The proposed ESM application and corresponding practitioner-centered data platform can be adapted to improve practitioner-family communications and treatment management for children with ASD in the U.S. (especially for rural and remote locations) and abroad. In the U.S. especially, long-term behavioral data are not readily available for practitioners and could be a useful tool in pharmaceutical trials and this proposal can serve as the proof of concept.
Ubicomp Lab was one of many partners in the Presidential Challenge Grant project to bring more mental health services to Milwaukee communities. Read more here!
Link to Open Access: The Marquette Palliative Care System
Abstract: Worldwide, major clinical barriers to effective palliative care are the absence of a. current data on symptom type and intensity; b. immediately accessible information on practical, affordable, and effective interventions; and c. self-sustaining systems to facilitate physician engagement in continuing palliative care. Our adaptable system, developed for Bangladesh and Nepal, addresses these barriers. A tele-home palliative care program facilitates home care, minimizes expense, and encourages efficient professional practitioner involvement employing two information technology tools: a patient cell-phone “app” on the Android platform with a 15-item symptom questionnaire and an Internet website with health information sections for patients/families and for clinicians. The physician section contains a guide for patient symptom review, clinical practice palliative care guidelines, secured patient demographics, medical summaries, and current and past symptom reports along with prescription-writing capability. The system is managed by a local organization that registers patients and their physicians, instructs patients on the free downloadable application and completes their demographic and medical summaries, and arranges collection of a modest fee. The organization also ensures regular physician/clinic visits by the patient or by a family member with a patient phone check-in, at least every 2 weeks.
Ubicomp Laboratory has partnered with the Milwaukee Health Department, the Medical College of Wisconsin, United Community Center, Marquette University and Carroll University for a project on screening for diabetic retinopathy. Ubicomp Lab is building the software framework for the project. Learn more here and here!
Md Kamrul Hasan, PhD candidate of the Ubicomp Lab presented a paper titled “SmartHeLP: Smartphone-based Hemoglobin Level Prediction Using an Artificial Neural Network” at the American Medical Informatics Association (AMIA), 2018 in San Francisco, CA, Nov 3-7, 2018.
Ubicomp Lab member, Jannat Tumpa received Student Scholarship to attend Grace Hopper Celebration (GHC) on Sep 26-28, 2018 at Houston, Texas. The Grace Hopper Celebration is the world’s largest gathering of women technologists organized by Anita Borg Institute in partnership with ACM which takes place every year to encourage more women in technology and to reduce the gender disparity in the tech field. Among 22,000 people attending GHC this year, only 867 students received the scholarship from across the globe. The experience of GHC helped Jannat to broaden her vision about her career by attending various keynote and technical sessions, to make lots of new networks in the career expo organized by 200+ sponsor companies and to know more about current ongoing research from the poster sessions.
Smartphone-based Human Hemoglobin Level Measurement Analyzing Pixel Intensity of a Fingertip Video on Different Color Spaces has been published in the Elsevier Journal Smart Health (Vol 5-6, January 2018, pp. 26-39). In this work by members of the Ubicomp Lab, RGB information of a 10-second fingertip video is used to covert into different color spaces including hue, saturation, value, lightness, and gray. Later, features are extracted from all the combinations of 10 different colors and applied to a Partial Least Squares algorithm. A prediction model is developed with data collected from patients, and achieves a reliable accuracy for detecting hemoglobin level. The article is available here.