MIT Jameel Clinic breast cancer risk prediction model MIRAI is validated on more than 1.5 million mammograms in 43 hospitals in 14 countries

Community Jameel welcomes the MIT Jameel Clinic's announcement that its breast cancer 5-year risk assessment model – MIRAI – has been validated on more than 1.5 million mammograms in 43 hospitals across 14 countries.

MIRAI is a deep learning model that can analyse a patient’s mammogram to accurately predict the patient’s risk of developing breast cancer in the next 5 years.

By assigning a personalised risk score to the mammogram, MIRAI helps clinicians determine when a patient should return for their next screening.

The ability to identify high-risk patients is critical to helping clinicians prioritise patients and save lives.

According to the World Health Organisation, breast cancer is the most common cancer in women, causing 670,000 deaths globally in 2022.

Moreover, survival is widely inequitable — nearly 80% of deaths from breast cancer occur in low- and middle-income countries.

The MIT Jameel Clinic is the epicentre of artificial intelligence (AI) and health at the Massachusetts Institute of Technology (MIT) and was co-founded by MIT and Community Jameel in 2018.

With support from the Wellcome Trust, the MIT Jameel Clinic has launched a hospital network to provide free access to an array of cutting-edge AI tools, empowering healthcare systems by accelerating the mainstream usage of AI tools on a global scale.

MIRAI’s pioneering technology was built by researchers from MIT and Mass General Brigham on a team led by Professor Regina Barzilay, faculty lead for AI at the MIT Jameel Clinic, and then-student Adam Yala.

Regina was inspired to build the tool after her own breast cancer diagnosis when she learned there were no clinically-available AI models that assess breast cancer risk.