Breast cancer screening is one of the most important tools doctors have for finding cancer early, when treatment is more likely to succeed.
In the United Kingdom, millions of women are invited to have breast screening tests every year.
These tests, called mammograms, are X-ray images of the breast that can reveal signs of cancer long before symptoms appear.
Now, a major new study suggests that artificial intelligence may help doctors detect more breast cancers while also reducing the workload of health care workers.
The research, published in the journal Nature Cancer, found that AI technology could increase breast cancer detection by more than 10% and reduce the workload for specialists by over 30%.
The study was carried out by scientists, doctors, and software developers from the University of Aberdeen, NHS Grampian, and Kheiron Medical Technologies, which is now part of DeepHealth Inc. The research focused on how AI could be used in real-world breast cancer screening programs.
Breast screening is already a large and complex medical service. In the UK, all women between the ages of 50 and 70 are invited to have a mammogram every three years. This national program results in more than two million mammogram examinations each year.
Currently, every mammogram in the UK screening program is examined by two radiologists. Radiologists are doctors who specialize in reading medical images such as X-rays and scans. Having two experts review each image helps reduce the chances of missing cancer. However, this process is time-consuming and places a heavy workload on the health care system.
Even with this careful approach, detecting breast cancer can still be difficult. Some cancers are very small or hidden in dense breast tissue, making them hard to see in mammogram images. As a result, about 20% of breast cancers may still be missed during screening.
Another challenge is that many women are called back for further tests even though they do not actually have cancer. When a mammogram looks suspicious, doctors may ask the woman to return for additional imaging or a biopsy. However, most of these follow-up tests do not lead to a cancer diagnosis.
In fact, for every five women recalled for further testing, only about one is found to have breast cancer. The other four women undergo additional procedures and experience worry and stress, even though they do not have the disease.
Because of these challenges, researchers have been exploring whether artificial intelligence could help improve the screening process.
In the new study, the research team evaluated an AI software tool called Mia. The tool was developed by Kheiron Medical Technologies and is designed to analyze mammogram images and highlight areas that may show signs of cancer.
The evaluation was part of a project called GEMINI, which stands for Grampian’s Evaluation of Mia in an Innovative National Breast Screening Initiative. The project was led by the University of Aberdeen and carried out in partnership with NHS Grampian.
The researchers tested the AI system in the screening of 10,889 women who participated in the NHS breast screening program in northeast Scotland. The goal was to see how the AI system could support radiologists in reading mammograms.
To fully understand how AI might work in clinical practice, the researchers tested seventeen different ways of integrating the technology into the screening process. These scenarios explored different roles for AI, such as acting as a second reader or serving as an additional safety check after doctors had examined the images.
The results were impressive. The researchers found that using AI could increase the number of cancers detected by 10.4%. Many of the cancers identified by the AI system were invasive and high-grade tumors, which are more aggressive forms of cancer that can grow and spread quickly.
Detecting these cancers earlier can make a major difference in treatment success. Earlier diagnosis often means that patients can begin treatment sooner, improving their chances of recovery.
Another important finding was that AI could dramatically shorten the time it takes to notify women about their screening results. In the current system, women may wait around two weeks to receive their results. With AI support, this time could potentially be reduced to just three days.
The study also found that AI could help reduce unnecessary recalls. By improving the accuracy of mammogram readings, the technology may prevent many women from being asked to return for additional tests that ultimately turn out to be unnecessary.
Reducing these extra appointments could lower patient anxiety while also saving valuable medical resources.
Dr. Clarisse de Vries, the lead author of the study and a lecturer in data science at the University of Glasgow, explained that the findings are important because they show practical ways AI could improve breast screening programs.
She noted that the research demonstrates how AI can help doctors detect cancer more quickly and accurately while also reducing the number of women who need to undergo unnecessary follow-up tests.
The study also showed that the best results came from combining human expertise with AI technology. In the most effective scenario, AI acted as a second reader that replaced one of the two human readers, while also providing an additional safeguard. This approach increased cancer detection while reducing the workload for doctors.
Health systems around the world are facing increasing pressure. Many countries have shortages of radiologists, and medical imaging workloads continue to grow as populations age. AI tools could help relieve some of this pressure by assisting doctors with routine image analysis.
Professor Gerald Lip, clinical director for breast screening in northeast Scotland, said the results show that AI can support radiologists rather than replace them. According to him, AI helps doctors find cancers that might otherwise be missed and allows specialists to focus their attention where it is most needed.
Despite the promising results, AI is not yet used routinely in the UK’s national breast screening program. The UK National Screening Committee previously stated that there was not enough strong evidence to recommend widespread use of AI.
The researchers believe this new study helps fill that evidence gap by providing high-quality data from real-world screening settings.
The work also lays the foundation for larger studies in the future. One upcoming project, called the EDITH trial, will expand the research to examine how AI can support breast screening programs across multiple sites in the UK.
Experts say studies like this are essential for understanding how new technologies can safely be integrated into health care systems.
Overall, the findings published in Nature Cancer suggest that artificial intelligence could play a valuable role in the future of breast cancer screening. By helping doctors detect cancer earlier, reduce unnecessary testing, and manage growing workloads, AI may help improve care for millions of women.
However, researchers emphasize that AI should be used as a tool to assist doctors rather than replace them. Careful testing, regulation, and continued research will be necessary before the technology becomes a routine part of medical practice.
Even so, the study offers a hopeful glimpse into how technology and medicine can work together to improve early cancer detection and patient care.
If you care about breast cancer, please read studies about how eating patterns help ward off breast cancer, and soy and plant compounds may prevent breast cancer recurrence.
For more health information, please see recent studies about how your grocery list can help guard against caner, and a simple way to fight aging and cancer.
Copyright © 2026 Knowridge Science Report. All rights reserved.
