AI mammograms could detect breast cancer earlier, new study finds

Boston 25 News
216 followers
10 Video Views·May 26, 2026
A new study suggests artificial intelligence could transform how breast cancer is detected — potentially catching the disease earlier than traditional screenings alone.
Find more at Boston25News.com: https://www.boston25news.com/news/local/ai-powered-mammograms-may-detect-breast-cancer-early-new-study-finds/J733NUGZRVGSRL5RZPD3VMV4WA/
Follow Boston25News on:
YouTube: https://www.youtube.com/@Boston25News
Facebook -- https://www.facebook.com/Boston25News
X -- https://x.com/boston25
Instagram -- https://www.instagram.com/boston25/
TikTok -- https://www.tiktok.com/@b25news?lang=en
Download the Boston 25 News app: https://www.boston25news.com/mobile-apps/
May is Women's Health Month, and it's meant to encourage women to prioritize their well-being
and also schedule their preventative screenings.
Catherine is back now to discuss how one of those screenings might someday change, Catherine, thanks to technology.
Well, women who are over 40 hear it all the time.
Don't forget to schedule your yearly mammogram.
Those tests are read by two radiologists, and they sometimes lead to more testing to rule out an area of concern.
But a large study out of Sweden suggests that process could someday become quicker and more accurate, thanks to AI.
I grew up in a house where my mother was dying of breast cancer.
Candio Terry was just 18 years old when she lost her mother to breast cancer.
She went on to become a beloved radio personality in Boston.
But beyond her work in the studio...
The fight against breast cancer became my own personal charity,
and I raised hundreds of thousands of dollars for the fight against breast cancer.
Then she became a patient.
When I was diagnosed with breast cancer, I was terrified.
Terry's diagnosis of stage 1 invasive breast cancer came during a routine mammogram, a test she'd had many times.
But that appointment was different.
I had a doctor who literally kept on sending me back in for more pictures, more pictures, more pictures.
And I remember thinking, she's concerned about something.
Candio Terry says the persistence of that doctor really left an impression on her,
the urge to keep testing, just on a hunch that there was still a diagnosis to be made.
But technology could soon offer a helping hand.
There's been a lot of interest in using AI to look at all different kinds of x-rays, particularly mammograms.
Dana-Farber medical oncologist Dr. Harold Burstein has been tracking a Swedish study of more than 100,000 mammography patients.
Half of their mammograms were read by two radiologists.
The rest were read by a single radiologist plus an AI program.
The AI module was able to detect actually slightly more cancers, so it missed fewer lesions.
And secondly, it did that without increasing the callback rate.
Dr. Burstein doesn't yet know how soon this method could be rolled out to patients.
And while he doesn't believe AI will ever fully replace radiologists,
he says it could be a boon to a field grappling with labor shortages.
It would begin to be possible to have a remote AI van or equivalent that could do a mammogram,
do a pretty good job at early detection of breast cancer and hopefully change the outcomes for women.
I think about my daughter. I think about my grandchildren.
And I think their road, should they ever go down it, will be paved so much more smoothly.
Oteri was treated with surgery and radiation.
She's in remission, and she's back to creating and thriving
thanks to the care she received and her early diagnosis.
And she believes patients shouldn't have one without the other.
Let's use AI-assisted mammography as a second set of eyes.
Let's use it as a tool to help us get there a little bit sooner.
But that personal touch that that doctor had that day,
that feeling that she had, let's check again, is irreplaceable.
Dr. Burstein says AI is already being used for early detection of breast cancer, but not during mammograms.
He says a local doctor developed a program that can better calculate a woman's risk of developing breast cancer,
potentially leading to earlier detection even before a mammogram is done.
