Mirai and Me: Promoting Breast Cancer Prediction in Nigeria Through AI Innovation
It was a sunny afternoon as I walked with classmates to Massachusetts General Hospital (MGH), excited for the MIT Jameel Clinic’s maiden summer bootcamp, where I discovered the wonders of AI in every class. It’s been over a year and I’m thrilled to be working with a team using Mirai, an AI tool developed at the Jameel Clinic, for predicting breast cancer up to 5 years ahead. In Nigeria, 18.1% of cancer deaths are from breast cancer, with an incidence of 54.3 per 100,000 and a survival rate of only 40%. Mirai has the potential to significantly improve these outcomes.
Working with Mirai has been enlightening, and Jameel Clinic’s summer bootcamp classes greatly aided my learning experience. Mirai analyzes 4 images for each patient and uses it to predict breast cancer risk over five years. At Nisa Premier Hospital Abuja, the retrospective dataset, generated using a GE Senographe Pristina 3D mammography device, is stored on a Picture Archiving and Communication System (PACS). There are plans to expand the image pool via a screening campaign. This can enhance diversity and further reduce any potential bias associated with Mirai.
Mirai offers hope for breast cancer detection and treatment, but there are challenges to its deployment in Nigeria. These include low awareness about Mirai among medical professionals, skepticism of the usefulness of AI in health, socio-cultural barriers, inadequate medical equipment, and limited mammograms due to infrequent screenings. Rural areas, home to 50% of Nigeria’s population, have limited access to healthcare services — only 12% have access to physicians and just 19% have access to nurses, worsening their access to screening opportunities.
Solutions include public awareness campaigns, increased healthcare funding, and specialized AI training for medical personnel, which this project actively supports.
A talk at the Jameel Clinic summer program inspired me to participate in this Mirai project in my home country to predict breast cancer and potentially save lives, as early detection is crucial in reducing cancer related deaths. In particular, I was motivated by a session at MGH on a deep learning model called “Sybil,” which is used to predict future lung cancer. That talk, which focused on radiology and cancer detection, was my favorite and left a lasting impression.
I’m reminded of my team’s presentation at the summer program on Viz.ai — a health company emphasizing the importance of swift action with the catch phrase, “Time is brain.” During strokes, every passing minute leads to the death of about 2 million brain cells, highlighting the critical role of AI in early detection. Similarly, early cancer detection can significantly improve survival rates.
For breast cancer, early detection can lead to a survival rate of up to 99%, and Mirai has the potential to make this a reality. This is why I am delighted to participate in this Mirai project, which leverages AI to enhance healthcare outcomes. Participating in this initiative is both fulfilling and is a meaningful application of what I learned at the Jameel Clinic summer program.
