by Tyler Wells Lynch
Anyone working on the frontiers of science will tell you: to advance the state of the art requires more than just domain expertise. For scientists like Sarah Ostadabbas, assistant professor of electrical and computer engineering at Northeastern University, it requires a collaborative, interdisciplinary approach.
She would know. As director of the Augmented Cognition Laboratory (ACLab) at Northeastern, Professor Ostadabbas works in computer vision, a subset of machine learning, and healthcare technologies. The two fields have remarkably different goals and techniques. While the former gives computers the gift of “sight,” the latter aims to revolutionize the healthcare domain by bringing advanced technologies to the diagnosis and treatment fronts.”
What’s the value in combining the two? In a presentation she’ll deliver at the Institute for Experiential AI’s (EAI) inaugural event on April 6, Professor Ostadabbas will explain just that, including details about her award-winning research into the use of learning algorithms to help diagnose childhood developmental disorders.
At ACLab, Professor Ostadabbas and her colleagues work on some of the most pressing problems in AI. Her research looks for ways to improve human information-processing capabilities through adaptive interfaces to augment the decision-making process in the healthcare domain. Recently, she received a career grant from the National Science Foundation. The award will allow her to study how computer vision and machine learning can be used to identify early signs of autism spectrum disorder (ASD) in infants.
How is this possible? Ostadabbas has developed advanced vision-based representation learning algorithms to model the motor behavior of infants in their natural environments, such as a crib or a playground. The goal is to uncover the complex, hidden relationship between physical movement and neurodivergent disorders like ASD that form later in life.
Ostadabbas began working on this project as a collaborative effort in 2019. To advance her studies, she hired a post-doc with a specialty in experiential AI — a human-centric approach to AI that combines the unique strengths of both humans and computers — and drew further collaboration from students, mentors, and researchers at a variety of universities and institutions.
Core to both Ostadabbas’ research and the vision of the Institute for Experiential AI more broadly is a multi-disciplinary approach that enables collaboration between domains. EAI believes this approach produces the kinds of insights and real-world applications needed to advance the state of the art. For Ostadabbas, it is a key reason why she joined the faculty at EAI and why she’s so excited to participate in its inaugural event.
“What we do at the institute is beyond just sharing the experience between academic settings and workplaces,” she says. “The experiential model has enabled the Institute to generate an environment that enables cross-domain collaboration. I am very excited to hear about some of the highlights of these activities and outcomes during the inaugural event.”