Dr Helen Frazer, the clinical director at St Vincent’s BreastScreen, has spent the past 10 years developing an AI tool called BRAIx. This new, innovative technology holds great promise to help radically improve the effectiveness of breast cancer screening. This innovative AI technology identifies women at high risk of breast cancer who may have been overlooked in traditional mammogram screenings. The researchers trained the tool on nearly half a million mammograms taken from Australian women, collected between 2016 and 2017. They validated this data with a separate dataset of nearly 4,500 Swedish women.
Frazer’s impetus for developing BRAIx comes from her direct experiences with cases where mammograms missed cancer. She hopes to find ways to make breast cancer detection much more accurate and help save more lives in the long run. This AI tool uses screening results from women to identify who are at high risk. It narrowly focuses on those who are in the highest 2 percent of that risk category. According to Brinker, these women have a 50-85% risk of developing breast cancer. Their probability is as high or higher than that of people with identified BRCA1 gene mutations. Alarmingly, among these high-risk women, one in ten went on to develop breast cancer. All of this was the case even though their mammograms were normal.
The Potential of BRAIx
He hopes that making artificial intelligence part of routine breast cancer screening will change the way the healthcare system operates. He sees a day where zero deaths from the disease are possible. Current screening methods have been criticized for being “one-size-fits-all,” or lacking the ability to take into account the nuances of individual risk factors. Using BRAIx really adds a level of precision to mammogram data analysis. This new and exciting technology has the potential to greatly increase early detection rates.
The algorithm by definition learns density, and it’s not confused or confounded by it,” said Frazer. Its digital operating abilities are at a pixel level whereas the resolution of the human eye is many times less than that. This powerful new functionality helps BRAIx look beyond the surface to detect more subtle indicators of risk that could easily be overlooked.
The adoption timeline for BRAIx aims for a national rollout in less than five years. Dr. Frazer argues that earlier screening, potentially starting at age 40 or younger, could significantly impact the overall survival rates of breast cancer. Ninety thousand Australian women are expected to lose their lives to breast cancer over the next 25 years. We need to do something about that,” she stated.
Addressing the Challenges
Dr. Frazer admits, even with progress, the healthcare system is still making it difficult. As the strain on hospital resources only increases, concerns for the potential neglect of patient safety and quality are evermore dangerous. Jess Armstrong, a healthcare professional who’s been helping inform these developments, was hopeful that AI can be woven into systems we already have.
Armstrong, defiantly responded, “Bring it on! Our hospital systems are under tremendous stress and human error is inevitable.” This AI tool rapidly reveals the bottom line and provides a score based on that. It speeds up wait times and assists you in determining if you require further scans.
Dr. Wendy Ingman, an expert in breast cancer research, further stressed the importance of BRAIx in improving detection methods. She pointed out that there are dozens more AI algorithms across the world already achieving success in predicting a woman’s risk for breast cancer. She thinks this one is by far the most to have progressed in defining what that risk is, and the specific level of that risk.
Future Implications
BRAIx is about much more than improving detection rates. It could truly revolutionize how we do breast cancer screening. As current methodologies prove less effective for certain demographics, there is an urgent need for innovative solutions like this AI tool.
As Dr. Frazer comments, broadening screening protocols, even if slightly, would increase the chances of catching these potential cases earlier. We don’t have the resources to test everyone. With AI, testing is more affordable and conducted quicker, giving us the opportunity to cast a wider net to younger populations,” she stated.
BRAIx has the potential to not just improve, but revolutionize the lives of individual patients. It’s got the potential to massively boost public health in the process. This innovative tool enhances risk assessment and allows for more proactive intervention. It can save lives and dramatically cut the number of women who die from breast cancer.
