EPICC Research Team

The goal of the EPPIC to investigate pain prediction from objective sensor and OMICS data using machine and deep learning methods, and precise pain interventions falls in line with my area of expertise, which broadly speaking is in mobile health and artificial intelligence methods for medical image analysis and smartphone sensing of user health. Specific to healthcare, since 2011, I have been working on an NSF/NIH-funded project to develop a mobile application for patients with advanced diabetes, which automatically analyzes the healing progress of their foot ulcers and helps them manage their condition at home. The work of EPPIC is closely related to that project. I worked with PhD students who researched wound image analysis algorithms using machine and deep learning and developed a smartphone application for that project. I have also worked on a range of other mobile applications including one for obesity counseling, for administering exercise as a drug to mitigate alcohol addiction, and for analyzing smartphone biomarkers for various ailments including Traumatic Brain Injury (TBI) and infectious diseases. I have also taught advanced courses in digital image analysis, computer graphics and mobile and ubiquitous computing for fourteen years. I have chosen to work with this strong interdisciplinary team as I believe the technical as well as clinical expertise will allow us to advance the study and treatment of chronic pain. The proposed research will utilize my expertise in leading large, collaborative research projects involving artificial intelligence, medical image analysis and mobile health research.

https://www.wpi.edu/people/faculty/emmanuel 

Dr. Lisa Conboy is the Director of Research at the New England School of Acupuncture at MCPHS University.  She is also faculty at NESA at MCPHS University and a faculty of Research Methodology.  In addition, Dr. Conboy has a research appointment at Beth Israel Deaconess Medical Center’s (BIDMC) Department of Gastroenterology Department and is an Instructor at the Osher Research Center at Harvard Medical School. 

Dr. Conboy is a social epidemiologist and a sociologist with over 20 years of research experience in Complementary and Alternative Medicine (CAM) such as acupuncture. She is interested in the associations between social factors and health. Most recently, her research focused on the use of acupuncture in the treatment of the complex medical illness Gulf War Illness.  Her Army-funded RCT found statistically significant reductions in pain and other symptoms.  She has been primary and co-investigator on 10 successful NIH projects through her positions at the BIDMC at Harvard Medical School, and Research Director of the New England School of Acupuncture at MCPHS University.

She is well published in the areas of Complementary and Alternative Medicine, Women’s Health, qualitative research methodology and complex systems science. She is also a founding member of the Kripalu research collaborative which examines the mental, physical, and spiritual benefits of yoga, meditation, Ayurveda and other holistic and mind-body therapies.

I am an expert in designing user-centered technologies that improve decision making and user experience. My research, which focuses on examining decision behavior and user experience, falls under the broader umbrella of human computer interaction (HCI) research. I study how computers can support human decision making and how humans experience computer use. I develop theoretical models that explain user experience and predict user behavior. I have expertise in qualitative research methods (e.g., observations, interviews) and quantitative methods (e.g., survey design/analysis, data analytics). In addition to subjective (e.g., surveys) and objective (e.g., performance) measures, I also use neurophysiological measures (e.g., eye tracking) to assess user experience, information processing, and decision behavior. 

My early research focused on improving decisions that were made using computerized tools. Because humans use their vision predominantly to process information, I decided to use eye tracking to study user experience and decision behavior. I secured private funding to establish a research program at WPI for developing tools, techniques, and methods that use eye movement data to advance information processing and decision making theories. Through this research program, my colleagues and I conducted work that shows examining pupillary responses in saccades and fixations separately can result in a more sensitive measure for information processing and decision behavior. Our work also shows that the compactness or density of a single fixation represents intense information processing, or focus. Based on these findings we developed a new concept for identifying focused attention and developed a method for capturing micro patterns of fixations (i.e., patterns of individual gaze points within a single fixation) in a gaze stream (US Patent App. 15/662,965). There are two reasons why the study of fixation micro patterns is particularly groundbreaking when contrasted with existing approaches for identifying fixations in gaze streams: 1) existing approaches do not account for the compactness of a fixation, and 2) existing approaches lack the sensitivity to discover the densest clusters of gaze points within a single fixation. These findings resulted in securing additional private funding to establish a research program for designing smart decision support technologies such as those that utilize neurophysiological signals to respond to user needs.

Because of the effectiveness of eye tracking in user research, I secured private funding to develop courses at WPI that incorporate eye tracking in understanding user cognition and behavior. This allowed eye tracking to be included in more student projects. For example, eye tracking was used in a number of student projects that examined the effectiveness of text simplification on cognitive effort and reading comprehension. The results of these student projects secured private funding from Amazon for the project “Automatic Text Simplification to Improve Text Comprehension of People with Cognitive Disabilities. The success of incorporating advanced eye tracking techniques in courses and student projects also resulted in a recent NSF grant for supporting the research and education program at WPI that focuses on designing user friendly robots assisting decision makers in workplace (NSF DGE 1922761). 

I have backgrounds in mental health services research, addiction psychiatry, and clinical trials, assessments of psychological symptoms and outcomes; design and implementation of behavioral interventions; and recruitment and retention. My research has focused on improving services for vulnerable populations. I have led and co-led several services research and clinical intervention studies funded by NIH, including as Co-PI on a CTSA-funded study of MBSR for low-income people of color (UL1TR000161 NIH/NCATS). I received my teacher training at UMass Medical School where I also served as the director of the MBCT program. Currently I am Director of MBCT training at the CHA Center for Mindfulness and Compassion. I have been leading MBSR and MBCT groups since 2011. My experience with successful collaborations has taught me that frequent communication among project members and a realistic research plan, timeline, and budget are essential.

I am currently an Associate Professor at University of Massachusetts Medical School (UMass). As the Associate Director of Research in the Department of Family Medicine, I have more than 15 years of experience as an Integrative Medicine clinician and mindfulness, health disparities, and chronic pain researcher. I am also the director of medical group visits in the Center for Integrated Primary Care at University of Massachusetts Medical School. My research concentration is patient-oriented research regarding chronic conditions, chronic pain, and non-pharmacological evidenced based medicine using both qualitative and quantitative methods. My current research is focused on the adaptive role of technology to support health behavior change and reducing symptoms using self-management. With funding by Patient Centered Outcome Research Institute (PCORI), I was the primary investigator on a comparative effectiveness randomized controlled trial of mindfulness medical groups visits compared to a primary care visit for participants with chronic pain. My research now focuses on the implementation of innovative technologies such as Embodied Conversational Agents and the GEMINI platform in low income diverse patient population including patients with chronic pain (NIH SBIR R43AT010460) or patients with cardiac risk factors using OWL for Hypertension (NIH CAPCAT U54HL143541 center grant).

As an NIH-funded researcher for over 20 years, former Vice Provost of Biomedical Research at UMass Medical School, and Peterson Family Dean of Arts and Sciences at Worcester Polytechnic Institute, my research remains a critical component of my responsibilities. Our lab uses various behavioral, neuroendocrine. and neuro-technologies to identify and characterize neuronal plasticity and behavior associated with stress and stressors including addiction, anxiety, and depression in preclinical models and clinical cohorts. In the last decade, our laboratory has shifted to a more translational approach and incorporates a strong clinical research component focusing on complimentary and integrative interventions, that are known to decrease stress like Mindfulness-based Stress Reduction (MBSR). The long-term goal of our work is to provide an understanding of the unique features of specific biomarkers and central mechanisms that regulate stress and build resilience. 

Recent clinical investigations have focused on assessing efficacy and identifying and characterizing the neural correlates of mindfulness. Our initial trial used a customized mindfulness protocol, “Keeping Weight Off,” to maintain weight loss. Our group documented a significant change in functional neural connectivity between the amygdala and ventromedial prefrontal cortex in the intervention group and changes in connectivity between the prefrontal cortex and several ROIs was associated with the change in depression symptoms at 6 months. This study provides preliminary evidence of neural mechanisms that may be involved in the impact of mindfulness on weight loss maintenance and/or depression that may be useful for designing future clinical trials, and predictive and mechanistic studies.  These and related data provide strong support of our current UH3 funded trial of a mindfulness-based intervention for hypertension, where I serve as MPI. I also co-lead a large, privately funded interdisciplinary collaborative project between WPI and Harvard’s McLean Psychiatric hospital which is focused on the use of machine learning to identify early predictors of severe depression and suicidality.  My combined research and administrative experience at both UMass, WPI, and as a current member of External Advisory Boards of two program projects I believe I have a strong foundation to lead this initiative to identify predictors of chronic pain, a major stressor for millions. 

https://www.wpi.edu/people/faculty/jaking

I am pursuing translational research on disease biomarkers and predictors of therapeutic responses, with the ultimate goal to match the right patient with the right treatment, particularly in multi-symptom diseases defined by patient reported outcomes such as pain and quality of life. In this context I have studied patients with Irritable bowel syndrome (IBS), inflammatory bowel disease (IBD) and Gulf War Illness (GWI). We employee multidisciplinary approaches integrating biological (ie genetic polymorphisms, cytokine levels, blood metabolites) and psychosocial parameters towards patient stratification, disease monitoring and assessment of treatment effectiveness. For example, in patients with IBS, we found that serum levels of osteoprotegerin correlated with clinical improvement overtime; and that TWEAK serum levels predicted clinical response. Furthermore, in the same study, we found that a polymorphism of the COMT gene, a regulator of the brain reward system, was associated with high responses to a placebo treatment. Those studies were supported by R01AT004662 “Omics and Variable Responses to Placebo and Acupuncture in Irritable Bowel Syndrome” in which I served as a PI. I have been collaborating with Dr Lisa Conboy of the EPPIC for more than a decade and I have served as a co-Investigator on her DOD-supported studies on GWI and the effectiveness of acupuncture treatment. I entered the Gulf War Illness field as a New Principal Investigator in 2016 and led the study W81XWH-16-1-0528 “GWI: Molecular Analysis of Disease Endophenotypes and Response to Acupuncture Treatment” with the aims to develop GWI diagnostic biomarkers and predictors of therapeutic responses using a targeted proteomics (SOMAscan) approach. The study has just been completed and we are in the process of publishing our findings. Preliminary data supporting the current application are derived from this study. I have also served as a co-Mentor in Dr Jacobson’s K01AT004916 “Structural Integration for Chronic Low Back Pain-cLBP” and as a co-Investigator in W81XWH-15-1-0640 “Novel Autoantibody Serum and Cerebrospinal Fluid Biomarkers in Veterans with Gulf War Illness”. As a member of the Gastrointestinal Research Group in our Institution (Beth Israel Deaconess Medical Center), I have maintained a Biorepository with samples (serum, plasma, total blood, DNA, RNA, mucosal biopsies) from a large number of well characterized patients (IBD, IBS, cLBP) as well as from matched healthy controls (HC). Chronic diseases like musculoskeletal pain with no established treatments pose an extraordinary scientific challenge and I am strongly committed to search for biomarkers that confirm diagnosis and provide information about the underlying disease mechanisms that can lead to more effective, personalized treatments. I am genuinely excited to participate in EPPIC, an innovative multidisciplinary and multilevel integrating initiative.  Along with an insightful team of investigators who have the expertise, background and proven record of publications and collaborations we are poised to carry out this project successfully.

I have 18 years expertise in development, assessment, and application of algorithms and tools to study structure, function, and evolution of macromolecular interactions through data integration. My current focus is in (i) developing integrated computational approaches that combine high-throughput interactions together with structural, and NGS data to understand how genetic and structural variations, gene dosage, and posttranscriptional modifications rewire the macromolecular interactions, and (ii) establishing the relationship of these effects to complex genetic disorders. The majority of current research projects in my lab leverage interactomics, structural genomics, NGS, and clinical genomics data obtained from public and private sources. I have also had a long track record of designing computational tools and databases that efficiently process and analyze massive volumes of biological and biomedical data. Serving as a Director of the Bioinformatics and Computational Biology Program at Worcester Polytechnic Institute and having served as a co-chair of University of Missouri Informatics Institute core initiative in translational bioinformatics and personalized medicine in the past, I focus on bringing together computational scientists, experimental biologists, and clinical researchers through projects like the proposed one. I have also had a number of collaborations on educational and research projects with data scientists, translational systems biologists, microbiologists, geneticists, and biostatisticians. During my career, I have successfully trained postdoctoral researchers as well as graduate, undergraduate, and high school students. Combined, my research expertise and previous federally funded experience in bioinformatics of disease and computational genomics, my leadership in translational bioinformatics and computational biology initiatives, and a strong-track record of interdisciplinary collaborations make me well-prepared for my role in this project.

I am excited to bring my experience and knowledge to the EPPIC team. The expertise I can provide is the innovative analytical methods that can be used to raise the analysis power and improve the findings for this proposal. I am a professor in biostatistics, and have been engaged in the study and support of scientific and analytic methods as applied to health outcomes for many years. I have extensive experience in providing analytical support for the proposals, for both clinical trials and observational studies. The methods I have developed and used in my collaborative work include machine learning techniques (supervised and unsupervised), predictive models, longitudinal and missing data models, omics analysis models, multilevel models, and survival models. I co-led several projects and carried out methodology development research on the longitudinal study of substance use, mental disorders, and hypertension control. My expertise and experience have prepared me to lead statistical modeling and data analysis in the proposed project. As PI or co-Investigator on several university, state, HRSA and NIH funded grants, I laid the groundwork for the proposed research by developing effective research designs and methods relevant to the prediction and forecasting of cardiovascular disease and mental disorders. In addition, I successfully administered the projects (e.g. staffing, research protections, budget), collaborated with other researcher teams, and produced several peer reviewed publications from each project.

My background in biopsychosocial (including omic) and imaging-based mechanisms and predictors of disease and collaborative project administration provides a strong foundation for my roles in the EPPIC.  I coordinate multisite mindfulness and MRI based clinical trials in my current positions at WPI and UMass Medical School.  My neuroimaging related expertise includes study design and multi-site coordination, data acquisition and analysis, anatomical and functional (both resting state and task based) imaging, and connectivity analysis in rodents, primates, and humans.  Work in my lab has used genetics, epigenetics, and fMRI to investigate the effects of early life social stress on neural connectivity in rodent models of social stress and depression and has expanded to related clinical studies of peripartum endocrine interventions, maternal depression and anxiety in migrant Latinas, and sex differences cognition and neural connectivity in non-human primates.  Current MRI projects include a NCCIH UH3 funded stage IIa RCT of the effects of mindfulness training on hypertension, where we are investigating the neural correlates of the mindfulness intervention in a collaboration with Dr. Eric Loucks at Brown University.  I am also involved in a project aimed at using machine learning to develop clinical and neuroimaging based predictors of depression and suicide in collaboration with McLean Psychiatric Hospital and Harvard University, as well as a similar projects using social media to predict depression and the prediction of mild cognitive impairment and Alzheimer’s from MRI data.

Evidence of the translational relevance of my previous studies is demonstrated by my current positions at WPI and UMass Medical School and related grants and publications.  A current EPPIC based project with Dr. Gardiner, Dr. Ruiz, and Dr. Rodriguez is identifying psychosocial mediators of pain trajectories in response to integrative medicine interventions.  A collaboration with Dr. Rodriguez and Dr. Hudson Santos at UNC Chapel Hill on postpartum depression in migrant Latinas has was awarded CTSA funding and already produced three substantial manuscripts.  Our initial publication reported DNA methylation of stress related genes in this high risk, underserved population that predicts discrimination exposure and suggests that ethnic discrimination acts by inducing neuroplasticity.  Additional analyses of these data revealed strong associations between ethnic discrimination, acculturation, and depression and anxiety, and these relationships were dependent on the methylation level of the gene for the glucocorticoid receptor.  Finally, due to the growing importance of the role of inflammation in psychiatric illness and behavior in general, we conducted mediation analyses which indicated that epigenetic markers of immunosuppression and inflammation mediate resilience and sensitivity, respectively, to the adverse effects of prenatal discrimination stress.  While our UH3 funded clinical investigation of neural correlates of a mindfulness intervention for hypertension is ongoing, we published a report of changes in neural connectivity in subjects in a mindfulness based clinical trial for weight loss, and also identified associations between connectivity and depression symptomatology from this RCT.   I am active in participating in collaborations with other labs, both basic and clinical, to maximize the productivity of my research projects and am skilled at organizing the work of these collaborations and obtaining related funding. 

I am an attending anesthesiologist and assistant professor within the Department of Anesthesiology and Perioperative Medicine at the University of Massachusetts Medical School. I specialize in pain medicine and I have been practicing in an academic environment for nearly 10 years. Currently, I serve as Chief of the Division of Pain Medicine in the Department of Anesthesiology and Perioperative Medicine at the University of Massachusetts Medical School and our clinical partner UMass Memorial Health Care. Academically, I teach several high demand inter-specialty rotations to medical students, residents, and fellows at the bedside daily. 

I have personally treated thousands of patients with a wide variety of chronic pain conditions including chronic low back and neck pain, complex spine disorders, CRPS I and II, various neuralgias, cancer pain, perioperative pain management in patients with opioid use disorder. As our clinical footprint has expanded with the addition of faculty to our division my clinical practice has become more focused on chronic pain related to disorders of the spine.  My current clinical template allows me to evaluate and treat 20-30 patients per day (Monday-Friday), most of whom suffer from chronic musculoskeletal pain.

I have spent most of my professional time building a high-quality clinical practice, teaching physician learners, and creating a team of specialist who care for a diverse community of patients. However, at this point I am extremity eager to work as a clinical partner with a strong interdisciplinary team of well-respected scientists with the goal of improving our ability to objectively analyze patient pain levels and measure their response to complementary treatments. Overall, my clinical expertise regarding the management of painful conditions of the spine, along with significant motivation to begin to formally contribute the scientific advancement of pain research, will allow me to be a successful collaborator for this project.

My rigorous training in biopsychosocial theory and methodology, including measurement and assessment of biomarkers of health and their relationships with psychosocial phenomena, give me the necessary background to meaningfully collaborate on this project. In my current position as an Assistant Professor of Psychological & Cognitive Sciences and Neuroscience at WPI, I conduct interdisciplinary research on the dynamic relationships among social factors and physiological processes. My expertise includes salivary cortisol, plasma biomarkers, remote data collection (including EMA methods), mHealth technology, wearable sensors, randomized controlled research, laboratory and field experimental designs, survey research, qualitative and interview methods, and various data analytic approaches. In addition to conducting research in my own lab, I actively pursue national and international research collaborations and am skilled at interfacing with and leading multi-site teams. Of particular relevance here, I am currently part of a multi-institution interdisciplinary collaboration investigating psychosocial mediators of pain trajectories in response to integrative medicine interventions.

I am a professor of Computer Science and of Data Science at WPI. My research is in Artificial Intelligence, Machine Learning, and Data Mining, and their applications to Medicine and Health. I have extensive experience working on multidisciplinary teams with clinicians on several domains including sleep, pain, stroke and obesity. My collaborators and I have developed novel, high-performing machine learning methods, including deep learning networks, for analyzing physiological medical data; and machine learning approaches for discovering patterns in behavioral data, aimed at transforming behaviors and improving health.

My research is in human-computer interaction, with an emphasis on the application of functional near infrared spectroscopy (fNIRS). One focus of my research is on next-generation interaction techniques, such as brain-computer interfaces, physiological computing, and reality-based interaction. I design, build and evaluate interactive computing systems that use machine learning approaches to adapt and support the user’s changing cognitive state and context. I also investigate novel paradigms for designing with accessibility in mind, particularly for the Deaf community. Much of my work also explores effective human interaction with complex and autonomous systems and vehicles. My work has applications in areas such as education, transportation, medicine, creativity support, gaming, and complex decision making.

My career is focused on computing systems, as a programmer and systems analyst early in my career and as a university professor over the last 30 years. My interests have focused on both the practical issues and the theoretical models that explain and predict effective deployment and use of computing and information systems – in organizations and by individuals. My methods expertise includes both qualitative methods (e.g., observations, interviews, and their coding) and quantitative methods (e.g., survey design and analysis, structural equation modeling, data science, and algorithm design).  In 2005 I turned my attention to healthcare information systems. I first focused on electronic health record (EHR) and personal health record (PHR) systems, and how their use leads to changes in healthcare delivery organizations such as those providing primary care.  Since 2010, my research has focused on personal health information technology (IT) applications on smartphones that are designed to engage patients in caring for themselves. One stream of this research developed an app focusing on patients with type 2 diabetes and associated diabetic foot wounds, funded by NSF (PI:Strong).  Like the current work of EPPIC, this included tracking physical activity of patients and collected data via the app, as well as interviews and short questionnaires.  The app was tested in the lab with patients and eye-tracking equipment, at the UMMS Wound Clinic, and in patient homes.

My research focuses on 1) developing mathematical and statistical methods to analyze neuroscience data, 2) developing methodologies and tools to characterize the relationship between the brain dynamics and behavior, and 3) utilizing these theories and models in clinical applications. In my research, I have worked to integrate methodologies related to signal processing, statistical inference, model identification, stochastic modeling, optimization, and control theory to develop more appropriate tools and techniques for the analysis of diverse forms of neural, behavioral, and other physiological signals. I have developed mathematical and analytical tools for multiple NIH and DARPA-funded projects. For instance, I have developed a toolbox called COMPASS, an open-source general-purpose software toolkit for computational psychiatry, which incorporates time-series analysis techniques for analysis of behavioral data. In a more recent research effort, I have worked on developing AI-backed games and cognitive experiments that help us to better study behavioral dynamics and its underlying neural mechanisms. I have also successfully produced several peer-reviewed publications from each of these projects.

Ruofan Hu is a PhD student in Data Science at WPI. Ruofan's research interests are in machine learning applications to medical domains, including pain medicine.

Wafaa is a third-year PhD student in Computer Science at WPI. Her study's objective is to develop a novel assessment tool that measures functional impairments and pain levels using machine learning to analyze IMU sensors data and demographic attributes of patients with Knee or Hip Osteoarthritis. Currently, she works on predicting the knee and hip pain and function level using the pain and physical function subscale questioners from the Knee Injury and Osteoarthritis Outcome Score (KOOS) and Hip Injury and Osteoarthritis Outcome Score (KOOS). Her work explores the feasibility of using machine learning on IMU data to identify and quantify mobility limitations in these patients.

Connor McLaughlin is a third-year computer science major at Worcester Polytechnic Institute from Atlanta, GA. His research interests include dimensionality reduction and image processing techniques applied to bioinformatics.