AI Shows Promise in Detecting Hard-to-Find Breast Cancers

United States: Artificial intelligence (AI) could be used to prevent such breast cancers that are found between routine mammograms, by picking up those that trained radiologists will fail to notice, a new study claims.

New study finds AI may reduce overlooked breast cancer cases by 30%

The use of AI in mammography may decrease the cases of interval breast cancers by 30 percent, researchers recently discussed in the Journal of the National Cancer Institute. Interval breast cancers are the ones that develop between screenings, as reported by HealthDay.

Researchers have noted that AI has the potential to detect tumors that radiologists failed to notice based on the visualization of mammograms. The program also might be capable of identifying those with faint or underlying signs arguably below the range of the human eye.

“This finding is important because these interval cancer types could be caught earlier when the cancer is easier to treat,” lead researcher Dr. Tiffany Yu, an assistant professor of radiology at the David Geffen School of Medicine at UCLA, said in a news release.

“For patients, catching cancer early can make all the difference,” she continued. “It can lead to less aggressive treatment and improve the chances of a better outcome.”

AI Detects Subtle Signs Radiologists May Miss

For the research, researchers examined data from close to 185,000 mammograms between 2010 and 2019, and drew attention to 148 cases of interval breast cancer.

The team went over the cases of interval breast cancer to try to understand why the tumors hadn’t been diagnosed sooner, and then used a commercially available AI program known as Transpara to see if it could find what the radiologists couldn’t.

Artificial intelligence (AI) can help prevent #breastcancers that develop between routine #mammograms, by catching ones that trained radiologists would overlook, a new study says. https://t.co/u3jC376pdM
— HealthDay News (@HealthDayTweets) May 14, 2025

More results indicated certain highlights:

Around three-quarters of mammograms that were initially identified as normal were later associated with an interval breast cancer.

In 90% of cases in which cancer was evident on the mammogram, the error was made by the radiologist in missing it.

89% of cancerous conditions have very feeble indications that may have been detected.

69 % of cancers that could not be seen on the mammogram in any way.

However, the AI was not good at identifying true interval cancers, which were absent at the time of screening, as it only identified roughly about 50% of them, a representation of the results.

“While we had some exciting results, we also uncovered a lot of AI inaccuracy and issues that need to be further explored in real-world settings,” senior researcher Dr. Hannah Milch, an assistant professor of radiology at the David Geffen School of Medicine, said in a news release.

AI’s Limitations Still Evident

“For example, despite being invisible on mammography, the AI tool still flagged 69% of the screening mammograms that had occult cancers,” Milch said. “However, when we looked at the specific areas on the images that the AI marked as suspicious, the AI did not do as good of a job and only marked the actual cancer 22% of the time.”

Newer studies are required to determine the use of artificial intelligence in breast screenings, researchers said.

Although AI is imperfect and should not be used alone, Yu said it has a promising future, as reported by HealthDay.

“It has the potential to be a good second set of eyes, especially for the kinds of cancers that are hardest to catch early, ” she said. It is basically about providing radiologists with better tools, which give patients a greater chance of detecting a cancer at its earliest stages, potentially resulting in more lives being saved.