BME Seminar: Peter Washington
Monday, October 17th, 2022 - 12:00 p.m.
Peter Washington
Assistant Professor of Information and Computer Sciences
University of Hawaiʻi at Mānoa
"Data Science for Digital Behavioral Phenotyping"
Keating 103 | Live Zoom, passcode: BearDown
(Instructor permission required for enrolled students to attend via Zoom)
Hosts: Dr. Beth Hutchinson and Dr. Shang Song
Persons with a disability may request a reasonable accommodation by contacting the Disability Resource Center at 621-3268 (V/TTY).
ABSTRACT:
Early childhood is the most efficacious opportunity to impact long-term health and learning. However, there are major bottlenecks to care which have resulted in a massive shortage of clinicians for diagnosis and treatment. These gaps in healthcare access disproportionately affect underserved and marginalized populations. Dr. Washington's talk discusses novel data science methods for developing a streamlined system for continuously phenotyping children with potential developmental delays by leveraging distributed non-expert crowd workers in conjunction with machine learning algorithms. This involves collecting diagnostically rich information from children and their parents in a secure and trustworthy manner, curating a reliable and capable crowd workforce for precisely extracting behavioral features, and training behavioral computer vision classifiers for automatic detection of neurodevelopmental concerns. Throughout Dr. Washington's talk, he will highlight the opportunities for further methodological data science innovations in quantifying human behavior more broadly.
BIO:
Dr. Peter Washington is an assistant professor in information and computer sciences at the University of Hawaii at Manoa in Honolulu, Hawaii. Dr. Washington completed his bioengineering Ph.D. at Stanford University, where he was a Stanford Interdisciplinary Graduate Fellow. Previously, Dr. Washington completed an MS in computer science at Stanford University and his BA in computer science at Rice University. Dr. Washington is interested in interdisciplinary data science problems at the intersection of health informatics, machine learning, and human-computer interaction. His long-term research goal is to make neuropsychiatric and mental healthcare more accessible and effective through the development of data science methods which span the entire data science pipeline: trustworthy data capture paradigms, precise data labeling algorithms, robust machine learning models, and real-world application of the resulting models and analyses. He also enjoys teaching data science.