Interior_Banner_Events

Artificial Intelligence Evaluation of Breast Tumor Morphology in Different Racial Groups

Robyn Abernethy, Emma Kate Banks, Marena Fleming, Soline McGee, Ellona Moulds, Mia Damiano, Alyssa Davenport, Savannah Finley, Megan Johnson, Heather Dunn, Marlee Marsh, Amber Stone, Department of Bioengineering, Department of Animal & Veterinary Sciences, Clemson University and Columbia College, Clemson University, Department of Bioengineering, Department of Animal & Veterinary Sciences, 138 Poole Ag Building, Clemson SC 29634 1301 Columbia College Dr. - Columbia, SC 29203-Biology department

This study aims to use machine learning and artificial intelligence to determine morphological characteristics in different racial groups associated with belligerent phenotypes of breast cancer. Existing literature suggests the incidence of aggressive subtypes of breast cancer has been linked to racial disparities. White women have a higher rate of developing breast cancer, yet lower mortality rate compared with African American women who are 42% more likely to die if diagnosed.There is a disparity in the age of diagnosis of breast cancer present between black and white women, where black women are diagnosed around 61 years of age and white women are likely to be diagnosed at a median age of 69 years.This information could be due to varying factors including genetic predispositions, socioeconomic status, access to adequate health care and/or physician bias throughout diagnosis and treatment. Scientists have determined that threatening forms of breast cancer are more common in younger African American women that reside in areas of low socioeconomic status. Despite advances in determining the genetics of breast cancer, previous studies have not investigated morphological differences in breast cancer tissue samples across racial groups. This study evaluated images of breast cancer tissue obtained from the Genomic Data Commons (GDC) data portal from the National Cancer Institute (NCI) database. Images were categorized by race, gender, stage, and type of breast cancer. Artificial intelligence (AI) computer programs were created to establish histological pattern variations of tumor samples when comparing different racial groups. Visual data and statistical analyses of cancer tissue will be presented and are anticipated to support different morphologies in breast tumor samples between black and white women. 




Additional Abstract Information

Presenters: Robyn Abernethy, Emma Kate Banks, Marena Fleming, Soline McGee, Ellona Moulds

Institution: Clemson University

Type: Poster

Subject: Microbiology

Status: Approved


Time and Location

Session: Poster 8
Date/Time: Tue 5:00pm-6:00pm
Session Number: 5651