AI Technology Predicts Personalized Five-Year Risk of Developing Breast Cancer

Breast cancer remains one of the most common cancers among women, with about one in eight receiving a diagnosis in their lifetime. Despite widespread use of mammography, about 34% of patients in the U.S. are still diagnosed at later stages when treatment is less effective. Current risk prediction methods rely on questionnaires that assess factors such as age, race, and family history, but these have limited accuracy. A new artificial intelligence (AI)-driven approach can now predict a woman’s five-year risk of breast cancer more accurately.

The software, called Prognosia Breast, was developed at Washington University School of Medicine in St. Louis (WashU Medicine, St. Louis, MO, USA) and licensed to Prognosia Inc., a startup company. It is a pre-trained machine learning system that analyzes mammogram images and age to generate a personalized five-year risk score. The technology is compatible with both 2D full-field digital mammography and 3D digital breast tomosynthesis, and its outputs are aligned with U.S. national risk reduction guidelines.

Trained on tens of thousands of mammograms from patients, the system learned to recognize early disease markers invisible to the human eye. Internal testing showed that it predicts breast cancer risk 2.2 times more accurately than standard questionnaire-based methods. The device produces an absolute five-year risk score that can be directly compared to national incidence rates, helping clinicians assess whether a patient’s risk is elevated.

The FDA has granted Prognosia Breast Breakthrough Device designation, expediting its path to full approval and clinical use. The designation recognizes its potential to significantly improve risk prediction and prevention. The software is designed to complement radiologists’ reviews, not replace them, and is intended to integrate seamlessly into existing mammography workflows.

The technology could be rapidly deployed, as infrastructure for mammography is already widespread and more than 75% of women aged 50 to 74 in the U.S. undergo mammograms every two years. Identifying high-risk women in advance will allow them to access preventive options such as more frequent screening, additional imaging like MRI, or preventive therapies. A clinical trial is being planned to combine the AI-derived risk score with current screening protocols to refine risk management strategies.

“We are excited about the potential of this technology to improve risk prediction and prevention of breast cancer broadly, no matter where a woman is getting screened,” said Graham A. Colditz, MD, DrPH, one of the developers of the software package. “The long-term goal is to make this technology available to any woman having a screening mammogram anywhere in the world. No matter the type of imaging they receive, our data show the software’s potential to identify women at increased risk of developing breast cancer over the next five years, providing them with opportunities to take targeted steps to reduce that risk.”

Visit expo >

Mobile X-Ray System

K4W

Digital X-Ray Detector Panel

Acuity G4

Ultrasonic Pocket Doppler

SD1

Computed Tomography System

Aquilion ONE / INSIGHT Edition