seed grant

Date: 2024-3-28 12:30 pm

Location: 
Technology Square Research Building (TSRB, 1st Floor Ballroom)
85 Fifth Street NW
Atlanta, GA 30308

Gian-Gabriel Garcia

garcia

Toward Fairer Type-2 Diabetes Diagnosis and Care: A Long-Term and Pipeline-Level View


Grant Participants: Gian-Gabriel Garcia (co-PI, ISyE), Jovan Julien (co-I, ISyE and Public Policy), Juba Ziani (co-PI, ISyE)

Lecture Presenter: Gian-Gabriel Garcia 

ABSTRACT
Type-2 Diabetes Mellitus (T2DM) is one of the most common chronic diseases in the United States, affecting about 10% of Americans. T2DM is irreversible, and when left untreated, can increase risk for several health complications, including nerve damage, heart disease, stroke, and negative mental health impacts. To this end, the early disease stages of T2DM, i.e., pre-diabetes, are reversible and the later stages are treatable. Accordingly, early screening, detection, and treatment are critical to reducing the rates of progression to T2DM and mitigating the adverse effects of T2DM among those who already have it. Yet, in the United States, T2DM can often go undetected until its later stages, with each missed detection stage leading to worsening health outcomes and increasing financial burden. Further, people from disadvantaged and underserved groups often face lower access to care, leading to more missed detection and greater downstream disease burden.  

In this research, our goal is to build a mathematical model to optimize investments across screening and treatment resources while reducing disparities across disadvantaged populations. To do so, we take a novel, zoomed-out approach that models the entire pipeline of disease management from screening to treatment, rather than treating each piece in isolation. Crucially, our modeling will consider how the prevalence, progression, screening, and treatment of T2DM disease across different subpopulations will impact future disease progression in these populations, along with their resource needs. This mathematical framework can inform health policy on how to allocate limited resources across a heterogeneous population and provide insight on modeling of interventions in other chronic diseases with similar progression characteristics to T2DM, thereby reducing disparities in long-term health outcomes across populations while lowering the need for costly–—and for many unaffordable—treatments over time. 

BIO
Gian-Gabriel Garcia is an Assistant Professor in the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Tech. His research involves the design, analysis, and optimization of data-driven frameworks at the intersection of prediction and decision-making as motivated by high-impact problems in health policy and personalized medicine. This research spans applications to diabetes, cardiovascular disease, concussion, opioids, and maternal health. He has secured federal funding as PI from the National Institutes for Health and the Agency for Healthcare Research and Quality. His research has also received recognition through various awards, including the IISE Transactions Best Paper Award in Operations Engineering and Analytics, INFORMS Bonder Scholarship for Applied Operations Research in Health Services, INFORMS Minority Issues Forum Paper Competition, and SMDM Lee B. Lusted Prize in Quantitative Methods and Theoretical Developments.  Before joining Georgia Tech, he was a postdoctoral fellow at Harvard Medical School. He earned his PhD and MS in Industrial and Operations Engineering from the University of Michigan and his BS in Industrial Engineering from the University of Pittsburgh. 

Andrea Jonsson

jonsson

What is the Role of the Human Voice in an Age of Technology


Grant Participants: Andrea Jonsson (School of Modern Languages), Stuart Goldberg (School of Modern Languages)

Lecture Presenter: Andrea Jonsson

Date: 2024-3-28 12:30 pm

Location: 
Technology Square Research Building (TSRB, 1st Floor Ballroom)
85 Fifth Street NW
Atlanta, GA 30308

ABSTRACT
As artificial intelligence, autotune, robotics, and voice technology platforms strive to create more realistic human-sounding interlocutors, it becomes increasingly important to problematize the role of voice as a cache of humanistic information that crosses interdisciplinary and transcultural boundaries.

Where technology can mimic cadence, linguistic syntax, and register in many languages, it is still the timbre of the voice that eludes recreation. Studying through a critical lens what is human about the human voice is necessary to enhancing our understanding of the limits of voice technology. This year, thanks to the IPaT:GVU research and engagement grant, I have undertaken an initiative to establish an innovative Interdisciplinary Voice Studies Lab called the Voice+ Research Lab housed in the School of Modern Languages that aims to explore the multifaceted nature of human voice expression endeavors to advance our understanding of the voice's role in human interaction, artistic expression, and social dynamics, with implications for fields ranging from linguistics and psychology to education and performance studies.

Through a combination of collaborative research, interdisciplinary discussions, and student and faculty engagement, the lab seeks to unravel the intricacies of vocal communication in various contexts, including speech, singing, and nonverbal vocalizations. Key objectives include investigating the physiological and psychological mechanisms underlying voice production, analyzing cultural and societal influences on vocal behavior, and developing novel methodologies for voice analysis and visualization. By fostering collaboration among experts from different fields, this project has started an invited speaker series and a student research symposium, created a website, and has plans to expand to include a podcast, a brown bag research presentation series, a cross-disciplinary reading group, and eventually an edited volume. 

BIO
Andrea Jonsson, associate professor of French and French program director at the Georgia Institute of Technology, received a Bachelor of Music Performance from McGill University and a PhD in French and Francophone Studies from the University of Pittsburgh. Andrea has several recent articles and chapters on voice, gender, and affect. Her co-authored book with Heather Warren-Crow (The University of Minnesota Press-Forerunners Series) is entitled Young-Girls in Echoland: #Theorizing Tiqqun. Her current book project, Amplified Intimacy: Voicing French Feminisms in Contemporary Pop Culture examines ways women use vocal intimacy to redefine soundscapes historically dominated by men in comedy, podcasts, performance, and music and is under contract with Liverpool University Press. 

Aviv Cohav

aviv cohav

Avery Gong

avery gong

What do people with blindness want in a robot dog guide?


Grant Participants: Bruce Walker (Psychology/Interactive Computing) and Clint Zeagler (IPaT)

Lecture Presenters: Aviv Cohav and Avery Gong

Date: 2024-3-28 12:30 pm

Location: 
Technology Square Research Building (TSRB, 1st Floor Ballroom)
85 Fifth Street NW
Atlanta, GA 30308

ABSTRACT
Dog guides provide an excellent mobility solution for blind or visually impaired (BVI) individuals. Unfortunately, these dogs must undergo a rigorous and costly training process, after which they only have about 8 “working years”. This results in a substantial problem with availability, and only a small percentage of BVI individuals who could benefit from having a dog guide are able to access them. To address this issue, we are developing a robot that could carry out the tasks of a traditional dog guide, with added capabilities. Our current work is focused on design research, with the goal of identifying specific functional and aesthetic design concepts to implement into a basic quadruped robot for an optimal blend of familiarity, innovation, and practicality. We have collected data through interviews and surveys to answer design questions pertaining to the appearance, texture, features, and method of controlling and communicating with the robot. These data inform us on directions for prototyping, enabling us to narrow in on designs that effectively meet the needs of BVI individuals.  

BIO
Aviv Cohav (he/they) is a MS student in Computational Science and Engineering at Georgia Tech. He is interested in research at the intersecting fields of Psychology, AI, and HCI, and intends to pursue a doctoral degree in Clinical Psychology.  

Avery Gong is a MS Computer Science student specializing in Machine Learning at Georgia Tech. She is interested in interdisciplinary research in HCI, AI, computer vision, and robotics. 

Aviv and Avery are involved in HCI research at the Georgia Tech Sonification Lab under the guidance of Dr. Bruce Walker. Their collaborative work centers on design research for the development of a robot dog guide for individuals with vision impairment.

Milka Trajkova

trajkova

Can AI Decode the Artistry of Human Movement?


Grant Participants: Milka Trajkova (School of Literature, Media, and Communication) 

Lecture Presenters: Milka Trajkova

Date: 2024-3-28 12:30 pm

Location: 
Technology Square Research Building (TSRB, 1st Floor Ballroom)
85 Fifth Street NW
Atlanta, GA 30308

ABSTRACT
As humanity stands at the brink of an artificial intelligence (AI) renaissance, a pivotal question emerges: Can AI decode the artistry of human movement?

Classical ballet is a critical piece of this question, given it is a synthesis of technical precision and profound creative expression. Many artistic movement domains are steeped in tradition but are ripe for innovation, and these invite us to explore the potential of AI to transcend the subjective and venture into the quantifiable realms of performance and wellness. To initiate this exploration, we are convening a summit from May 30-31, 2024, to begin a conversation on the way we transform the subjective assessments of artistic human performance into objective, quantifiable processes. Our workshop, hosted by the Georgia Institute of Technology, brings together researchers from industry and academia to harness AI to enhance human performance and wellness, with the goal of addressing the challenge of quantifying artistry.  

The potential ramifications for this work extend well beyond dance. As sports analytics have reshaped our understanding of athletic prowess, a similar approach to dance could redefine our comprehension of human movement, with implications for healthcare, construction, rehabilitation, and even aging as we know it.

BIO
Milka Trajkova, Ph.D., is a Research Scientist at the Georgia Institute of Technology working in the Expressive Machinery Lab. Milka leverages her past professional ballet career to explore how we can design non-invasive AI-based tools to enhance artistic human movement performance toward the democratization of training, learning, and creativity.  

Michael Cross

cross

Paula Gómez Z.

gomez

Artificial Intelligence based Abstract Review Assistant (AIARA)


Grant Participants: Michael Cross (RSI, CIPHER Lab, GTRI), Paula Gomez (SRE, CIPHER Lab GTRI), Mark Riedl (Professor and Associate Director of the Georgia Tech Machine Learning Center, School of Interactive Computing)

Lecture Presenters: Michael Cross and Paula Gomez Z.

Date: 2024-3-28 12:30 pm

Location: 
Technology Square Research Building (TSRB, 1st Floor Ballroom)
85 Fifth Street NW
Atlanta, GA 30308

ABSTRACT
The peer-review system, which relies on ad honorem subject matter experts (SME), is in crisis due to the decreasing number of reviewers since the COVID-19 Pandemic, doubling the time required for a blind review process. This situation creates a need for new tools and methods for conducting peer reviews. As the peer review process is well structured and includes an analysis of a set of components (title, keywords, IMRaD structure, length, clarity, and innovation), a peer review is an excellent candidate for AI training, addressing topics such as duplicate submissions, self-plagiarism, incomplete abstracts, IMRaD evaluation, and standardization of scores for the final selection of abstracts. This research proposes to implement such as Natural Language Processing (NLP) and Large Language Models (LLMs) for prototyping an opensource LLM, trained with information acquired through the Computational Design Conference organizations. The outcome will be a set of automatically produced reviews that will be evaluated and scored by SME. The long-term goal is the implementation of this AI reviewer in systems such as OpenConf. 

BIO
Michael Cross is a Research Scientist and IT Operations Branch Manager at the IT division of the Cybersecurity, Information Protection, and Hardware Evaluation Research (CIPHER) Laboratory, Georgia Tech Research Institute’s (GTRI). He has been Principal Investigator for Subject Matter Assistance and Resource Tool for Intelligent Engineering (SMARTIE), an IRAD aimed at bringing AI capable tools to researchers working with proprietary information. His experience includes End-User Computing manager, focusing on important principles in Human-Factors and Ergonomics to Information Technology Support Management. 

Dr. Paula Gómez Z. is a Senior Research Engineer at the RISC Unit at the CHIPER Lab, GTRI, where she leads the Systems Modeling research thrust, including spatiotemporal modeling, information and knowledge modeling, systems modeling, and MBSE. She obtained her Ph.D. in Computational Design from Georgia Tech, sponsored by Fulbright. She is instructor on the Vertically Integrated Project (VIP) program, Principal Investigator on research projects for Spatiotemporal Model of COVID-19 spread in buildings, and was MBSE task lead for the MBSE Gates Foundation Generation 2 Reinvented Toilet (G2RT) and the NASA Kennedy Space Center Visitor Complex PZ-interactive flooring system. Dr. Gomez’s work has been recognized by her peers with several international awards. She currently serves as Vice President of International Affairs for SIGraDi (2024-2026), and has served on the International board since 2017, as well as several scientific reviews committees, including eCAADe, SIMAUD, and ASCAAD. She was guest editor in chief of the International Journal of Architectural Computing, IJAC (2018-2024).