Artificial intelligence may detect cancer diagnoses much earlier. A recent study in the journal Radiology reported that AI accurately predicted one-third of breast cancer cases up to two years before they were diagnosed. The study analyzed imaging data and screening details from BreastScreen Norway exams conducted between January 2004 and December 2019.
The study found that women who were eventually diagnosed with breast cancer through these exams received an AI risk score from a “commercially available AI system.” These scores were categorized as 1-7 for low-risk malignancy, 8-9 for intermediate risk, and 10 for high-risk malignancy.
AI Achieves a Breast Cancer Breakthrough
The study evaluated AI scores and mammographic features, including calcifications and breast density, in a total of 2,787 screening exams involving 1,602 women with an average age of 59.
The findings showed that over 38% of screening-detected and interval cancers had a high AI risk score of 10 before breast cancer diagnosis.
(Interval cancers are those found between regular mammogram screenings.)
Among screening-detected cancers with AI scores available four years prior to diagnosis, 23% had a high-risk score of 10.
Solveig Hofvind, a study co-author and the head of the Norwegian Breast Cancer Screening Program, as well as a professor of radiography at Oslo and Akershus University College of Applied Sciences in Norway, expressed her thoughts on the findings.
“We were taken aback by the results, suggesting that a significant number of cancers can be identified even earlier, offering the potential for less aggressive treatment, reduced side effects, and improved quality of life,” she conveyed in an email to Fox News Digital.
New Research Reveals Potential Serious Side Effects of a Breast Cancer Drug
Dr. Brian Slomovitz, the director of gynecologic oncology at Mount Sinai Medical Center in Miami Beach, Florida, found the study “very interesting.” He was not part of the research team.
“In this, there is certainly a possibility for early detection, though not necessarily prevention,” he stated in an interview with Fox News Digital.
He also noted, “It’s crucial to remember that this is a retrospective study.”
Slomovitz emphasized the need to replicate these findings in a prospective manner if we intend to implement this in clinical practice.
Nevertheless, he acknowledged that the study is highly convincing.
As oncologists, we understand that the most effective approach to cancer is either prevention or early detection,” he stated.
“And if we can improve our ability to detect cancers at an earlier stage using artificial intelligence, it will result in better outcomes for our patients.”