MIT AI Model May Help Eliminate Racial Disparities In Breast Cancer Screenings
Photo Credit: Screening possible cancer illness in female breasts

MIT AI Model May Help Eliminate Racial Disparities In Breast Cancer Screenings

As artificial intelligence becomes more advanced, people are beginning to look into its medical possibilities. Recently, a team from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and Massachusetts General Hospital (MGH) announced the development of AI to predict breast cancer.

The AI model can help detect breast cancer up to five years in advance, Engadget reported. That alone is significant, but the researchers also claimed that their method helps eliminate racial disparities.

“Rather than taking a one-size fits all approach, we can personalize screening around a woman’s risk of developing cancer. For example, a doctor might recommend supplemental MRI screening for women with high model-assessed risk,” MIT professor Regina Barzilay, a breast cancer survivor and senior author of the team’s paper in Radiology said.

Researchers trained the AI model by showing it mammograms and known outcomes from over 60,000 patients treated at MGH. By doing so, the AI was trained to learn “subtle patterns in breast tissue that are precursors to malignancy.”

In addition, the AI model was found to be equally accurate for white and Black women. That’s huge because the Center for Disease Control and Prevention noted that although Black and white women get breast cancer at about the same rate, Black women die at a higher rate. Part of the issue is that white women are likely to get diagnosed earlier than Black women.

“Breast cancer mortality is approximately 40% higher among black women compared with white women, with faster decreases in mortality among white women,” one CDC study said.

“It’s particularly striking that the model performs equally as well for black and white people, which has not been the case with prior risk assessment tools,” Allison Kurian, associate professor of Medicine and Health Research and Policy at Stanford University School of Medicine, said. “If validated and made available for widespread use, this could really improve on our current strategies to estimate risk.”

The researchers hope to make their AI model part of standard care.