Exploring Gender Disparities in Time to Diagnosis

Screen Shot 2021-03-31 at 7.31.25 PM.png

Video


Team Information

Team Members

  • Tony Sun, PhD Candidate, Department of Biomedical Informatics, Graduate School of Arts and Sciences

  • Jennifer Chen, MA Student, Department of Biomedical Informatics, Graduate School of Arts and Sciences

  • Oliver Bear Don't Walk, PhD Candidate, Department of Biomedical Informatics, Graduate School of Arts and Sciences

  • Harry Reyes Nieva, PhD Candidate, Department of Biomedical Informatics, Graduate School of Arts and Sciences

  • Jaan Altosaar, Postdoctoral Officer of Research, Department of Biomedical Informatics, Graduate School of Arts and Sciences

  • Faculty Advisor: Noémie Elhadad, Associate Professor, Department of Biomedical Informatics, Columbia University

Abstract

Gender based disparities contribute to different healthcare outcomes. We propose a large scale per-phenotype analysis of gender differences in time to diagnosis across a longitudinal diagnostic process. Evaluating binary diagnosis classifiers and gender recall gaps across diagnosis timesteps, we observe gender differences in diagnosis accuracy across time, and through our per-phenotype measure of gender bias called Mean Squared Discrimination observe gender bias in TTD across phenotypes.


Contact this Team

Contact: Tony Sun (use form to send email)

Previous
Previous

Supervised Topic Modeling for Predicting Chemical Substructure from Mass Spectrometry

Next
Next

UltraProcessed Food Proxy Gene Signature Associated with Inflammation in Patients with Crohn's Disease