Utilization of an AI-Powered Chatbot for Enhancing Oral Cancer Awareness among African Americans

Presentation Time: 10:30 AM - 10:45 AM

Abstract Keywords: Education and Training, Clinical Trials Innovations, Machine Learning, Generative AI, and Predictive Modeling
Primary Track: Data Science/Artificial Intelligence
Programmatic Theme: Harnessing the Power of Large Language Models in Health Data Science

This study presents the development and expert evaluation of an AI-driven chatbot aimed at enhancing oral cancer education, especially among African Americans. Initial results indicate good usability and accuracy, though improvements are needed in user guidance and language simplification. A forthcoming Randomized Controlled Trial will assess the chatbot’s effectiveness against conventional educational materials, with findings expected by December 2024.
Speaker(s):
Nour Abosamak, MD
Virginia Commonwealth University

Author(s):

Outpatient Portal Use and Blood Pressure Management during Pregnancy

Presentation Time: 10:45 AM - 11:00 AM

Abstract Keywords: Patient-centered Research and Care, EHR-based Phenotyping, Measuring Outcomes
Primary Track: Clinical Research Informatics
Programmatic Theme: Digital Health Technologies for Patient Research

We investigated the association between systolic and diastolic blood pressure (BP), and outpatient portal use during pregnancy. We used electronic and administrative data from our institution. We categorized patients into two groups: (<140 mmHg systolic; <90 mmHg diastolic), and out-of-range (≥140 mmHg systolic, ≥ 90 mmHg diastolic). Random effects linear regression models examined the association between mean trimester BP levels and portal use, adjusting for covariates. As predicted portal use increased, both systolic and diastolic BP levels decreased for the out-of-range group. These differences were statistically significant for patients who were initially out-of-range. For the in-range group, systolic and diastolic BP levels were stable as predicted portal use increased. Results provide evidence to support a relationship between outpatient portal use and physiological outcomes during pregnancy. More research is needed to expand on our findings, especially those focused on the implementation and design of patient portals for pregnancy.
Speaker(s):
Athena Stamos, BS
The Ohio State University

Author(s):
Athena Stamos, BS - The Ohio State University; Naleef Fareed, PhD MBA - The Ohio State University Dept Biomedical Informatics;

Feasibility Assessment of a Wearable App to Manage Symptoms of Postural Orthostatic Tachycardia Syndrome Using Real-Time Heart Rate Monitoring

Presentation Time: 11:00 AM - 11:15 AM

Abstract Keywords: Mobile Health, Wearable Devices and Patient-Generated Health Data, Patient-centered Research and Care, Advanced Data Visualization Tools and Techniques
Primary Track: Data Science/Artificial Intelligence
Programmatic Theme: Digital Health Technologies for Patient Research

Postural Tachycardia Syndrome (POTS) is a chronic condition characterized by orthostatic intolerance and a significant rise in heart rate upon standing. Patients often experience debilitating symptoms, such as brain fog and chronic fatigue, which hinder daily functioning. Non-pharmacological management strategies, particularly pacing, are crucial for reducing symptom fluctuations and improving quality of life. Heart rate monitoring plays a key role in effective pacing, enabling patients to plan activities and prevent severe symptom onset. Recent technological advancements have increased interest in wearable devices for managing chronic conditions. This study examines the feasibility of using wearable technology to support symptom management in POTS patients. Through an Exploratory-Descriptive Qualitative approach, five key themes emerged, including personalized management strategies and the beneficial impact of real-time feedback. The findings suggest that wearable devices can enhance self-management, improve communication with healthcare providers, and empower patients to take a more proactive approach to their care.
Speaker(s):
Aileen Gabriel, Bachelor of Science
Department of Biomedical Informatics, University of Utah

Author(s):
Te-yi Tsai - University of Utah; Christian Reategui Rivera, MD - University of Utah; Patricia Rocco, DPT, MS - Department of Biomedical Informatics, University of Utah; AREF SMILEY, Assistant Professor/PhD - The University of Utah; Clayton Powers, DPT - Department of Physical Therapy, University of Utah; Jeanette Brown, MD, PhD - Department of Internal Medicine, University of Utah; Joseph Finkelstein, MD, PhD - University of Utah;

Enhancing AI Health Coach with LLM-powered Advanced Memory Framework

Presentation Time: 11:15 AM - 11:30 AM

Abstract Keywords: Machine Learning, Generative AI, and Predictive Modeling, Natural Language Processing, Mobile Health, Wearable Devices and Patient-Generated Health Data
Primary Track: Data Science/Artificial Intelligence
Programmatic Theme: Harnessing the Power of Large Language Models in Health Data Science

The integration of large language models (LLMs) into AI health coaches offers significant benefits, enabling more personalized, accurate, and context-aware health guidance. A key component of such an AI health coach is a memory mechanism that maintains a comprehensive conversation history between the user and the AI. This mechanism is essential for delivering personalized coaching plans, tailored health advice, and ensuring continuity in future dialogues. However, incorporating the entire conversation history into future prompts can exceed the token limit of LLMs, potentially diluting crucial information in the current dialogue. To address this challenge, we propose a novel memory framework that efficiently summarizes historical conversations into structured, time-variant tables and uses SQL queries to extract relevant information for future interactions. We present a comprehensive design and evaluation framework for memory writing, retrieval, and integration. Additionally, we demonstrate that a fine-tuned, quantized LLM can achieve performance comparable to cloud-based models (e.g., GPT-4), paving the way for both cloud and on-device LLM-based memory module deployment. The questionnaire-based user study also shows that users strongly prefer responses generated by health conversational agents with an integrated memory framework, highlighting improvements in personalization, comprehension, and naturalness, which significantly enhance user satisfaction and engagement.
Speaker(s):
Yikuan Li, M.Sci
Northwestern University

Author(s):
Hanyin Wang, PhD - Merck; Yuan Luo, PhD - Northwestern University;

Achieving and Maintaining Euglycemia During Pregnancy for Type 2 Diabetes Through Technology and Coaching: Usability Study

Presentation Time: 11:30 AM - 11:45 AM

Abstract Keywords: Mobile Health, Wearable Devices and Patient-Generated Health Data, Patient-centered Research and Care, Stakeholder (i.e., patients or community) Engagement
Primary Track: Clinical Research Informatics
Programmatic Theme: Digital Health Technologies for Patient Research

We conducted usability testing with five patients and eight care team members on a mobile Health application (care team and patients) and dashboard (care team). These digital tools are designed to support Medicaid-insured pregnant individuals who have type 2 diabetes (T2D) to better manage health-related social needs, T2D, and pregnancy. We built prototypes and collected feedback and future recommendations on functionality and user experience. These suggestions will guide future improvements of the digital tools.
Speaker(s):
Naleef Fareed, PhD MBA
The Ohio State University Dept Biomedical Informatics

Author(s):
Athena Stamos, BS - The Ohio State University; Christine Swoboda; Yiting Wang; Naleef Fareed, PhD MBA - The Ohio State University Dept Biomedical Informatics;

A Method for Enabling Digital Health Technologies in Clinical and Translational Research at an Academic Medical Center

Presentation Time: 11:45 AM - 12:00 PM

Abstract Keywords: Mobile Health, Wearable Devices and Patient-Generated Health Data, Implementation Science and Deployment, Clinical and Research Data Collection, Curation, Preservation, or Sharing
Primary Track: Clinical Research Informatics
Programmatic Theme: Digital Health Technologies for Patient Research

Federal- and state-level governance as well as local institutional oversight is changing rapidly to address the accelerated growth in the usage of digital health technologies (DHT)--such as apps, wearables, and websites—to enable clinical and translational research. While studies have described frameworks for assessing and/or implementing individual DHTs, to our knowledge there are none describing how to implement and support multiple DHTs at an academic medical center. A multi-disciplinary review process including information technology, institutional review board, legal, privacy and others identified 33 items to evaluate as part of onboarding studies using DHTs. In a one-year period, 80 review requests have been done for research (75) and non-research (5) usages. The 33-item evaluation list described in this study may be valuable to researchers and practitioners in other settings seeking to scale institutional support for DHTs.
Speaker(s):
Cindy Chen, MA
Weill Cornell Medicine

Author(s):
Thomas Campion, PhD - Weill Cornell Medicine;