Interests:
Natural language processing, machine learning, community engagement, Indigenous knowledge, epistemological justice, social drivers of health, ethics, intersectionality, and fairness.
Background:
Dr. Bear Don’t Walk is an Assistant Professor in Biomedical Informatics and Medical Education with a PhD from Columbia University in Biomedical Informatics, and an MA and BS from Stanford University. He has experience with training and evaluating machine learning models using structured and textual data, working with Indigenous communities to guide research and application, interrogating bias in clinical models, and instructing courses on machine learning and programming. Their work focuses on developing and building machine learning models for patient information extraction, identifying and leveraging social drivers of health in clinical informatics, community engagement, and incorporating Indigenous knowledge into biomedical informatics.
Current projects:
- Collaboratively Describing Community-Informed Social Drivers of Health for Patients Living with HIV: From Patient Perspectives to the Electronic Health Record
- Participating in a U01 grant funded research to explore how people experience hallucinations through mobile device data collection and machine learning
Teaching:
BIME 585: Professional Development In Biomedical Informatics
Other Roles:
Center for Indigenous Health Special Responsibilities
Currently accepting new students.
Representative publications:
Bear Don’t Walk OJ, McLester-Davis LWY, Trinidad SB (2025) Rectifying Genocidal Data Stewardship: A Commentary on Ethical and Legal Obligations for Sharing Data With Tribal Entities. Journal of Medical Internet Research 27:e77946
Bear Don’t Walk OJ, Pichon A, Reyes Nieva H, et al (2024) Contextualized race and ethnicity annotations for clinical text from MIMIC-III. Sci Data
Bear Don’t Walk OJ, Paullada A, Everhart A, Casanova-Perez R, Cohen T, Veinot T (2024) Opportunities for incorporating intersectionality into biomedical informatics. Journal of Biomedical Informatics 154:104653
Bear Don’t Walk OJ, Sun T, Perotte A, Elhadad N (2021) Clinically relevant pretraining is all you need. J Am Med Inform Assoc 28:1970–1976
Bear Don’t Walk OJ (2022) Inferring Race and Ethnicity from Clinical Notes: Annotation, Model Auditing, and Ethical Implications. https://doi.org/10.7916/mst0-g412
Feller DJ, Bear Don’t Walk IV OJ, Zucker J, Yin MT, Gordon P, Elhadad N (2020) Detecting Social and Behavioral Determinants of Health with Structured and Free-Text Clinical Data. Appl Clin Inform 11:172–181