Emory University and Georgia Tech Award AI.Humanity Seed Grants

Emory University and Georgia Tech have announced the inaugural recipients of $100,000 in seed funding from their collaborative AI.Humanity program.

AI.Humanity is an extension of the existing partnership between the two universities forged through Emory’s Constructive Collisions programming earlier this year. The grant recipients will use the funding to spur new research collaborations and expand existing partnerships leveraging artificial intelligence (AI) to improve society and daily life. Projects may incorporate research including but not limited to aspects of ethical and social considerations, social justice, health disparities, or bias in AI data.

“These teams are partnering to revolutionize AI and promote equity and improvement of the overall quality of human life,” said Deborah Bruner, senior vice president for Research at Emory University. “This is an exciting time for research departments at Emory and Georgia Tech. Congratulations to each of the winning teams!”

"This opportunity leverages our universities' combined strengths in AI, health, and cognition research,” said Rob Butera, vice president for Research Development and Operations at Georgia Tech. “Emory and Georgia Tech have a long history in collaborating, especially in biomedical research, and AI.Humanity pushes these collaborations into new domains."

The winning proposals were selected from a pool of more than a dozen entries across the two universities. The recipients of the AI.Humanity seed grants are:

AI Forest: Cognition in the Wild

Marcela Benitez, Emory University, College of Arts and Sciences, Department of Anthropology
Jacob Abernethy, Georgia Tech, School of Computer Science

In the proposed study, Benítez and Abernethy plan to develop and implement “smart” testing stations for long-term cognitive assessment and monitoring of wild capuchin monkeys at the Taboga Forest Reserve in Costa Rica. These testing stations will rely on AI and deep learning to recognize and track wild monkeys in real-time, allowing for targeted behavioral assessment and cognitive testing. The stations will also provide a novel method for long-term monitoring of cognitive abilities in wild animals. In doing so, the team will achieve an unprecedented level of control in a wild environment, providing opportunities for several studies linking cognitive performance to natural behaviors and ultimately overall fitness.

Applying Machine Learning Techniques to Improve Epidemiological Models Accounting for Urban Infrastructure Networks, Human Behavioral Change, and Policy Interventions

Lance Waller, Emory University, Rollins School of Public Health, Department of Biostatistics and Bioinformatics
John Taylor, Georgia Tech, Department of Civil and Environmental Engineering

This project will examine novel infectious diseases, which can be dangerous and require rapid public health response but can be challenging to model, especially in the early stages of a potential major outbreak. The team’s research focuses on characteristics of urban infrastructure networks (e.g., transport networks), which add density to and alter the order and structure of contact networks, often accelerating local disease transmission in the event of widespread infectious disease. The team proposes extending epidemiological models to incorporate the complex role of local differences in contact networks and the dynamic nature of human-human and human-infrastructure interaction networks in shaping disease transmission, human behavioral change, and policy interventions within metropolitan areas. Their project’s goal is to provide more accurate results than homogeneous mixing models and remain computationally feasible for guiding rapid policy decisions. 

Diabetic Ulcer Computational Sensing System

Marcos Schechter, Emory University, Department of Medicine
Rosa Arriaga, Georgia Tech, School of Interactive Computing

This project will explore computational approaches to detect changes in diabetic foot ulcers through models that analyze and interpret heterogeneous data and provide AI-driven interfaces that connect patients and clinicians. The team’s proposed human-centered computational sensing system will bridge current gaps and address the clinical challenge of automating wound screening and monitoring by characterizing ulcer severity and wound progression and predicting wound healing and recurrence. Additionally, the team will focus on underserved and minority communities to promote technologies to reduce disparities. This pilot proposal will enroll people from underserved communities at Grady Memorial Hospital, a public hospital where over 250 people are hospitalized with diabetic foot ulcers annually.