April 1, 2016
Making Sense of Big Health Data through Visual Analytics and Machine Learning
Amar K. Das, MD, PhD
Associate Professor of Biomedical Data Science, Psychiatry, & The Dartmouth Institute for Health Policy and Clinical Practice
Director, Division of Biomedical Informatics, Department of Biomedical Data Science
Director, Biomedical Informatics Core, SYNERGY Clinical and Translational Science Institute
Head, Informatics Collaboratory for Design, Development, and Dissemination
Geisel School of Medicine at Dartmouth
About the Presentation:
The speed by which we can generate and gather Big Health Data from health systems, social media, high throughput technologies and other sources has quickly outpaced our ability to use that complex data effectively and rapidly. We are in need of new data science methods by which health data consumers, such as policy makers, healthcare providers and patients, can find meaningful, actionable information. In this presentation, Dr. Das will discuss research in the Social Computing & Health Informatics Lab (SCHI Lab) to develop new visual analytics and machine learning approaches for Big Health Data that is deep, longitudinal, and heterogeneous. Dr. Das will demonstrate the application of our methods to breast cancer care and relate these efforts to mental health care and behavioral research as well.
March 4, 2016
The Ethics of mHealth
Dr. Tiffany Cvrkel
Bioethicist, Philosopher, and Lecturer
Molecular, Cell, & Developmental Biology
About the Presentation:
As the capacities and wide-spread acceptance of mHealth devices grow, so too does their potential in research and clinical contexts. Collecting and working with this type of data raises unique ethical concerns. The purpose of this talk is two-fold. First, we will be outlining some of the most pressing ethical challenges presented by mHealth. Second, we will start thinking about solutions, exploring how these ethical concerns can and should inform our use of this technology.
Deputy Director for User Interface Design
UI/UX Design; Information Visualization; Persuasive Communication
Lorie’s research resides at the intersection of UI/UX design, information visualization and persuasive communication. Her work uses design and technology to communicate in order to change people’s behavior and draw attention to new ways of seeing the world. In addition to her work at Dartmouth Lorie is the President/Co-Founder of TellEmotion, Inc. TellEmotion motivates people to conserve resources by creating an emotional connection to real-time use data. Lorie comes from an arts background. She worked on animated films and television programs that won many awards (two Emmys and the Cine Golden Eagle), and have been screened at the Museum of Modern Art, the Sundance Film Festival, the NY Film Festival, the London Film Festival and the Whitney Biennial.
Lorie is the Executive Director and Co-Founder of the Digital Arts Leadership and Innovation (DALI) Lab in the Computer Science Department at Dartmouth. At the DALI Lab, students design and develop technology tools to help our partners communicate effectively and maximize impact. Lorie lives at Cobb Hill Co-Housing where she is a beekeeper, gardener, and expert wood stacker.
R Goswami, T Tregubov, L Loeb. FrameShift: shift your attention, shift the story. SIGGRAPH 2015: Studio, 2015
L. Loeb, E Tice, T Tregubov. Emotionally engaging students to change behaviors and conserve resources. International Journal of Environmental, Cultural, Economic, and Social Sustainability 2010
E Tice, T Tregubov, L Loeb. GreenLite Dartmouth: Unplug or the polar bear gets it. SIGGRAPH 2009
C Bregler, L Loeb, E Chuang, Hi Deshpande. Turning to the Masters: Motion Capturing Cartoons. ACM Transactions on Graphics (TOG). Vol. 21. No. 3. ACM, 2002.
Saeed Hassanpour, PhD
Assistant Professor, Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth College
Machine Learning; Data Mining; Natural Language Processing
Saeed Hassanpour is an Assistant Professor in the Department of Biomedical Data Science at Geisel School of Medicine at Dartmouth College, with adjunct appointments in the Computer Science and Epidemiology Departments. His research is focused on developing new computational methods to capture and organize clinically meaningful information from complex and massive amounts of biomedical data. His lab uses this distilled information to provide intelligent tools to help biomedical researchers understand their data better and assist clinicians in medical diagnosis and practice.
Dr. Hassanpour completed his postdoctoral training in the Department of Radiology at Stanford University School of Medicine, where he developed novel computational methods to extract clinically significant information from radiology reports. Before his postdoctoral position at Stanford, he worked as a Research Engineer at Microsoft for more than 2 years. His research at Microsoft was focused on high-throughput semantic text analysis to extract user intents from Web search queries. He received his PhD in Electrical Engineering with a minor in Biomedical Informatics from Stanford University.
Saeed Hassanpour, Curtis P. Langlotz, “Information Extraction from Multi-Institutional Radiology Reports”, Artificial Intelligence in Medicine, 66:29-39, 2016.
Saeed Hassanpour, Curtis P. Langlotz, “Predicting High Imaging Utilization Based On Initial Radiology Reports: A Feasibility Study of Machine Learning”, Academic Radiology, 23(1):84-89, 2016.
Saeed Hassanpour, Curtis P. Langlotz, “Unsupervised Topic Modeling in a Large Free Text Radiology Report Repository”, Journal of Digital Imaging, 29(1):59-62, 2016.
Saeed Hassanpour, Martin J. O’Connor, Amar K. Das, “Clustering Rule Bases Using Ontology-Based Similarity Measures”, Journal of Web Semantics: Science, Services and Agents on the World Wide Web, Vol 25, pp 1-8, 2014.
Saeed Hassanpour, Martin J. O’Connor, Amar K. Das, “A Semantic-Based Method for Extracting Concept Definitions from Scientific Publications: Evaluation in the Autism Phenotype Domain”, Journal of Biomedical Semantics, 4:14, 2013.
MySafeRx™: An Integrated Mobile Platform for Buprenorphine Adherence
Funding Source: NIDA - National Institute on Drug Abuse (R34DA040086)
Funding Period: 04/15/2016 – 03/31/2019
Principal Investigator: Zev Schuman-Olivier, MD
Other project staff: Jacob Borodovsky; Qays Munir; Jackson Steinkamp; Lisa Marsch, PhD; Hai Yi Xie, PhD; Mark McGovern, PhD;
Opioid use disorder represents a serious public health issue. Rates of overdose deaths have been rising among young adults. Additionally, acute Hepatitis C (HCV) infection rates are increasing fast among young adults, especially in rural areas. Buprenorphine/Naloxone (B/N) is a partial opioid mu-receptor agonist dosed daily that prevents opioid withdrawal, blocks opioid euphoria and can prevent opioid overdose. Extended treatment with B/N increases rates of abstinence and retention in outpatient treatment and decreases risk of transmitting HCV. Younger adults have difficulty stabilizing, achieving abstinence, and remaining active in office-based opioid treatment (OBOT) with B/N as compared to older adults. Retaining young adults in B/N treatment is important because dropout is linked with relapse and overdose death. Low levels of adherence to buprenorphine during treatment is associated with dropout from treatment, therefore an intervention that supports medication adherence among young adults during periods of instability during OBOT may offer a critical mechanism for improving treatment outcomes among young adults with opioid use disorder.
MySafeRx™ is a novel integrated mobile platform for medication adherence. This innovative solution has the potential to be a new gold standard for remote adherence monitoring and diversion prevention. The MySafeRx™ platform, which was designed specifically for use with controlled substances, such as B/N, during addiction and mental health treatment, focuses on simultaneously enhancing medication adherence, preventing medication diversion, and providing daily recovery coaching support during periods of clinical instability (e.g. patients with opioid use disorder receiving OBOT with B/N). MySafeRx™ uses a mobile application that provides the capacity for daily remote supervised self-administration of B/N by integrating confidential text messaging, daily videoconferencing with visual observation of medication taking, and secure, electronic medication dispensers (i.e., Medicasafe 3000 device). The MySafeRx™ remote check-in coaching encounter, provides relational recovery coaching and motivational enhancement, during a critical daily window around the time of medication taking. Finally, by providing regular reports to prescribers through a dedicated prescriber web interface, MySafeRx™ increases prescribers' awareness of the level of medication adherence and recovery stability.
This study aims to assess feasibility of a 6-week course of the MySafeRx™ integrated platform provided for a population of vulnerable young adults (<34 years old) with opioid use disorder, enrolling patients during B/N induction and established B/N patients with recent illicit opioid use. The study aims to evaluate both patient and prescriber levels of acceptability and usability, and assess treatment effects on evaluation of patient stability over time. Finally, this study also aims to evaluate initial clinical efficacy of MySafeRx™ compared to standard care based on weekly TLFB self-reports of B/N adherence and on percent negative weekly opioid urine toxicology (during 6 weeks of trial).