Deep learning model analyzes mammograms to predict breast cancer risk up to five years in advance, now being tested at National Cancer Center Hospital
The MIT Jameel Clinic has launched a collaboration with Japan’s National Cancer Center Hospital to evaluate Mirai, an artificial intelligence tool that predicts breast cancer risk from mammography images. The deep learning model can assess a patient’s likelihood of developing the disease up to five years before diagnosis.
Breast cancer remains the most common cancer among Japanese women, accounting for roughly 23% of all female cancer cases. Approximately 98,782 women receive breast cancer diagnoses each year in Japan, with an estimated 16,000 deaths annually. When detected early, the five-year survival rate exceeds 90%.
The collaboration was announced at a ceremony attended by Mohammed Jameel, founder and chairman of Community Jameel, and Dr. Yasuyuki Seto, director of the National Cancer Center Hospital. Community Jameel cofounded the MIT Jameel Clinic in 2018 with the Massachusetts Institute of Technology to advance machine learning applications in healthcare.
Mohammed Abdul Latif Jameel on AI and Cancer Care
“Mirai is a powerful tool that harnesses AI to improve cancer care for women around the world,” Mohammed Jameel said. “With the Jameel family’s deep and longstanding connection to Japan, we are delighted that the MIT Jameel Clinic and Community Jameel are collaborating with the National Cancer Center Hospital to open the way for Mirai to improve care for Japanese women at risk of breast cancer.”
The study will analyze mammography data collected between 2013 and 2024 from individuals screened at the National Cancer Center Hospital and Yotsuya Medical Cube. Mirai will generate risk scores for each patient across a one-to-five-year horizon, then researchers will compare these predictions against actual outcomes to assess accuracy within a Japanese clinical context.
Mirai has already been validated on more than two million mammograms across 72 hospitals in 23 countries. The tool was developed at the Jameel Clinic, which serves as the epicenter of artificial intelligence and health research at MIT.
Personalized Screening Approach
Current Japanese guidelines recommend mammography screening every two years for women aged 40 and above. While mammograms can detect small lumps and microcalcifications invisible through self-examination, interpretation relies primarily on visual assessment and clinical experience.
Dr. Kan Yonemori, director of the Department of Medical Oncology at the National Cancer Center Hospital, said the study has the potential to contribute to improving women’s health by predicting individual breast cancer risk using large-scale mammography screening data. “Our team in Japan is committed to advancing this work as part of an international collaboration with the MIT Jameel Clinic, and we look forward to contributing meaningfully to this important global research effort,” he said.
Regina Barzilay, AI faculty lead at the MIT Jameel Clinic, expressed hope the research would inspire new approaches to breast cancer screening and treatment in Japan. Barzilay has led development of several clinical AI tools at the Jameel Clinic, which has become known for breakthrough discoveries including the antibiotics halicin and abaucin.
If validated, the findings could enable risk-based screening programs in Japan. Higher-risk individuals might receive more frequent monitoring, while those at lower risk could avoid unnecessary tests. This personalized approach contrasts with the current one-size-fits-all screening schedule.
Mohammed Jameel, who holds an honorary knighthood from Queen Elizabeth II for philanthropic activities, has built extensive partnerships between his family’s organizations and leading research institutions. Community Jameel supports four major centers at MIT, including the Abdul Latif Jameel Poverty Action Lab, whose co-founders won the 2019 Nobel Prize in Economics.
The Jameel Clinic represents one of several health-focused initiatives supported by the Jameel family. Community Jameel also cofounded the Jameel Institute with Imperial College London, which led critical modeling of COVID-19 spread during the pandemic.
Japan maintains particular importance for the Jameel family through decades of business relationships. Abdul Latif Jameel, the diversified network of businesses chaired by Mohammed Jameel, has partnered with Toyota Motor Corporation for 70 years as the authorized distributor in Saudi Arabia.
The current study receives support from Community Jameel and Jameel Corporation. Researchers will evaluate whether Mirai’s predictive capabilities, demonstrated in Western populations and other international settings, translate effectively to Japanese women, who may have different breast density patterns and cancer risk profiles.
National Cancer Center Hospital has operated since 1962 as a leading institution in cancer treatment and research. The hospital conducts industry-sponsored trials, investigator-initiated trials, first-in-human trials and international collaborative clinical trials.
Terms of the collaboration were not disclosed. Results from the mammography analysis are expected to inform future decisions about implementing AI-supported screening programs in Japan.
