2023 Health Equity Data Challenge

Presented by AIM-AHEAD Data Science Training Core and Nightingale Open Science

May 23 - June 23, 2023

Challengers specified time to improve their models.
All teams will be scored and placed on a Leaderboard.
Finalists will be identified based on Leaderboard.
All Challengers must make solutions public on Github.

High Risk Breast Cancer Prediction

AIM-AHEAD DSTC and Nightingale developed this Data Challenge to introduce underrepresented researchers and trainees to AI/ML and Health Equity with the analytical goal to improve the prediction of the cancer stage in minority patient’s cancers above a baseline score, using the whole slide images generated by a breast biopsy.

Breast cancer is a leading cause of cancer death in women. From a routine mammogram, some women go on to have an invasive biopsy to better examine a concerning area. Since the 1990s, there have been advancements in surgical procedures and chemotherapy. But, known health disparities remain among breast cancer patients due to race, ethnicity, socioeconomic status, geography and access to equitable healthcare. This year, the Health Equity Data Challenge was developed to consider AI/ML solutions for analyzing breast biopsy image data and to demonstrate health disparity research to an underrespresented audience in biomedical research.

Contains whole slide images from 1,000 breast biopsies, in 842 patients, from 2014 to 2020. Minority populations found in the dataset include: Black or African American, American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, Hispanic or Latino.

Teams can include up to five challengers. Underrepresented populations in the U.S. biomedical, clinical, behavioral and social sciences research enterprises are strongly encouraged to participate. Challengers must be 18 years of age or older and be legal U.S. residents. CLICK FOR MORE INFO ON UNDERREPRESENTED POPULATIONS

Challenge finalists will receive travel awards to present their results at the AIM-AHEAD Annual Meeting in Washington, DC, August 1-3, 2023. Finalists will also receive an opportunity to publish their best solution. Additional awards of cash and compute credits maybe available.