Validation of an AI-Based Mammography Model (Mirai) for Early Breast Cancer Risk Prediction in Saudi Women
1
Nursing. Collage, King Abdulaziz University, Jeddah, Saudi Arabia
Popul. Med. 2026;8(Supplement Supplement 1):A2859
ABSTRACT
BACKGROUND:
Breast cancer (BC) is the most common cancer among women globally, with an estimated 2.3 million new cases reported in 2020 [1,2]. In Saudi Arabia, BC is also the most frequently diagnosed cancer among women, affecting approximately 27 out of every 100 [3]. Although mammography is considered the gold standard for early detection, its effectiveness is limited by false-positive results, overdiagnosis, and low adherence to regular screening [4,5]. These issues can lead to delayed diagnoses, unnecessary interventions, increased anxiety, and higher healthcare costs. Additionally, behavioral, cultural, and systemic barriers further reduce screening participation among Saudi women, contributing to late-stage detection and poorer outcomes [5-8]. To overcome these challenges, personalized screening strategies are essential. One promising solution is Mirai, an AI-powered, mammography-based deep learning model developed at the Massachusetts Institute of Technology (MIT). This model has demonstrated high accuracy and sensitivity in identifying women at elevated risk of breast cancer across several international populations [9,10]. However, its performance has not yet been validated in Saudi Arabia.
AIM:
To validate the predictive accuracy of the Mirai AI model using mammogram data from Saudi women.
METHODS:
Ethical approval and data-sharing agreements with MIT have been obtained. Data collection is ongoing at King Abdulaziz University Hospital. This retrospective cohort study included 6,000 mammograms images from women aged 45–70 with no history of BC and at least five years of consecutive screening. Each mammogram must include all four standard views (bilateral CC and MLO). The model’s performance will be assessed using AUC, C-index, and paired DeLong’s test. Final results are expected by June 2026. Conclusion & Implications: Early findings suggest that Mirai may enable BC risk prediction up to three years in advance. Its clinical integration may enhance early detection, reduce mortality, and improve screening effectiveness in Saudi Arabia.
Breast cancer (BC) is the most common cancer among women globally, with an estimated 2.3 million new cases reported in 2020 [1,2]. In Saudi Arabia, BC is also the most frequently diagnosed cancer among women, affecting approximately 27 out of every 100 [3]. Although mammography is considered the gold standard for early detection, its effectiveness is limited by false-positive results, overdiagnosis, and low adherence to regular screening [4,5]. These issues can lead to delayed diagnoses, unnecessary interventions, increased anxiety, and higher healthcare costs. Additionally, behavioral, cultural, and systemic barriers further reduce screening participation among Saudi women, contributing to late-stage detection and poorer outcomes [5-8]. To overcome these challenges, personalized screening strategies are essential. One promising solution is Mirai, an AI-powered, mammography-based deep learning model developed at the Massachusetts Institute of Technology (MIT). This model has demonstrated high accuracy and sensitivity in identifying women at elevated risk of breast cancer across several international populations [9,10]. However, its performance has not yet been validated in Saudi Arabia.
AIM:
To validate the predictive accuracy of the Mirai AI model using mammogram data from Saudi women.
METHODS:
Ethical approval and data-sharing agreements with MIT have been obtained. Data collection is ongoing at King Abdulaziz University Hospital. This retrospective cohort study included 6,000 mammograms images from women aged 45–70 with no history of BC and at least five years of consecutive screening. Each mammogram must include all four standard views (bilateral CC and MLO). The model’s performance will be assessed using AUC, C-index, and paired DeLong’s test. Final results are expected by June 2026. Conclusion & Implications: Early findings suggest that Mirai may enable BC risk prediction up to three years in advance. Its clinical integration may enhance early detection, reduce mortality, and improve screening effectiveness in Saudi Arabia.
| eISSN: | 2654-1459 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
