AI Tool BRAIx Improves Breast Cancer Risk Prediction, Offers Path to Personalized Screening

A novel Australian-developed artificial intelligence (AI) tool, called BRAIx, is demonstrating significant promise in improving the accuracy of breast cancer risk prediction, even in women with normal mammograms. The tool, detailed in a study published in The Lancet Digital Health, could pave the way for more personalized screening pathways and earlier detection of the disease.

How BRAIx Works

BRAIx utilizes deep learning algorithms to analyze mammographic images and calculate an individual breast cancer risk score. Unlike traditional risk assessment methods that rely on factors like age and family history, BRAIx extracts risk signals directly from the mammogram image itself. This allows for a more precise estimation of a woman’s probability of developing breast cancer within the four years following the screening. The risk score ranges from 0 to 99.9.

Study Findings

Researchers at the University of Melbourne based their study on data from over 96,000 women who participated in the BreastScreen Victoria screening program, as well as data from nearly 4,500 Swedish women [1]. The study revealed that women in the top 2% of BRAIx risk scores had a 9.7% probability of receiving a breast cancer diagnosis within four years – a risk comparable to, or even exceeding, that of some individuals with BRCA gene mutations [2]. The AI identified women at the same or greater risk of breast cancer as those with BRCA gene mutations.

Potential Impact on Screening

The development of BRAIx could shift breast cancer screening from a one-size-fits-all approach to a more personalized model. Women identified as being at higher risk could be offered additional tests, such as magnetic resonance imaging (MRI) or contrast-enhanced mammograms, for earlier detection. Conversely, women with very low risk scores might be able to extend the intervals between screenings, reducing unnecessary testing and anxiety [3].

BRAIx Development and Future Outlook

The BRAIx project is a collaborative effort focused on transforming breast cancer screening through improved analysis and interpretation of mammograms for Australian women [4]. Dr. Helen Frazer, clinical director at St Vincent’s BreastScreen, has been working on this technology for ten years and believes it’s a “breakthrough discovery” that the algorithm can identify signals for risk that the human eye may miss [2]. Researchers anticipate the tool will be ready for rollout within the next five years, but emphasize the need for further prospective trials and clinical evaluations to determine the best way to integrate it into existing screening programs.

Key Takeaways

• BRAIx is an AI tool that predicts breast cancer risk by analyzing mammograms.
• It provides a personalized risk score between 0 and 99.9, estimating the likelihood of developing breast cancer within four years.
• The tool has shown promise in identifying high-risk women who might be missed by traditional screening methods.
• BRAIx could lead to more personalized screening pathways, with targeted testing for those at higher risk.
• Further research is needed to fully integrate BRAIx into clinical practice.

With approximately 2.3 million breast cancer cases diagnosed globally each year, and over 20,000 diagnoses annually in Australia, advancements like BRAIx offer hope for earlier detection, improved treatment outcomes, and a reduction in the burden of this disease.