New AI tool predicts breast cancer recurrence faster

A new artificial intelligence tool may soon change how doctors understand breast cancer risk, offering faster answers and more precise guidance at a critical moment in care. The technology could help determine whether a patient’s cancer is likely to return, a question that shapes treatment decisions for millions.

Researchers at New York University developed the test by combining digital images of tumor tissue with basic clinical information. Their findings show that the system can match or even outperform widely used genetic tests.

Breast cancer treatment has advanced in recent decades. Many patients now survive and go on to live full lives. Still, recurrence remains a serious concern.

Doctors often face difficult choices when deciding how aggressively to treat the disease. Some patients receive chemotherapy that may not be necessary. Others may need more treatment than they receive.

“Breast cancer is not a single disease, and decisions about how aggressively to treat it are often difficult,” said Krzysztof J. Geras, who led the study.

To guide these decisions, clinicians often rely on genomic tests. These tests analyze gene activity in tumor tissue to estimate the risk of recurrence. While helpful, they come with limits.

They can take weeks to deliver results. They are also expensive and require specialized processing that uses up valuable tissue samples.

The new approach builds on something doctors already use every day. When cancer is diagnosed, tissue samples are placed on glass slides and examined under a microscope.

The AI system analyzes these same slides in digital form. It looks for patterns that may not be visible to the human eye.

It then combines those findings with clinical details such as tumor stage, patient age, and hormone receptor status. The result is a score that estimates how likely the cancer is to return.

“This research shows that an AI test can read the same tumor slides pathologists already examine and, combined with basic clinical details, accurately estimate how likely a patient’s cancer is to return,” Geras said.

The system uses a method called self-supervised learning. Instead of relying only on labeled examples, it first studies large amounts of data to learn patterns on its own.

“The model’s accuracy doesn’t come from hand-labeled data alone,” said Yann LeCun. “It comes from self-supervised pretraining that lets it learn rich representations first, which then translate into strong downstream performance, a recipe that should generalize far beyond breast cancer and, more broadly, is the kind of new AI science these hard problems demand.”

This approach allows the model to detect subtle features that may otherwise go unnoticed. It also makes it more adaptable to different types of data.

To evaluate the system, researchers analyzed data from more than 3,500 patients. These patients came from 15 groups across seven countries.

The model showed strong performance in separating patients into higher and lower risk categories. It used standard statistical measures to confirm its accuracy.

One key metric, known as the concordance index, showed that the system reliably ranked patients by risk. Another measure, the hazard ratio, confirmed clear differences between risk groups.

The AI test also performed well across different types of breast cancer. This includes triple-negative and HER2-positive cases, where existing genomic tests often fall short.

This broad performance suggests the system could help a wider range of patients.

One of the most immediate benefits is speed. Traditional genomic tests can take weeks to complete. The AI system can deliver results in a matter of hours once slides are digitized.

This faster turnaround could reduce stress for patients waiting for answers. It may also help doctors make treatment decisions more quickly.

Cost is another advantage. Because the system uses existing slides, it avoids the need for additional laboratory processing. This could make testing more accessible.

It also preserves tissue samples. Unlike genomic tests, which consume part of the sample, the AI approach leaves material available for future use.

The system also offers clearer risk categories. Current genomic tests often place patients into low, intermediate, or high risk groups. The intermediate group can be difficult to interpret.

The AI model provides more distinct separation. This may help doctors make more confident decisions about treatment.

For patients, this could mean avoiding unnecessary chemotherapy or receiving more targeted care when needed.

The system also showed strong results over different timeframes. It was able to predict outcomes several years into the future, including the risk of distant recurrence.

The study highlights a broader shift in medicine. Instead of relying on one type of data, researchers are combining multiple sources to better understand disease.

By merging imaging and clinical information, the AI test provides a more complete picture of each patient’s situation.

It also shows how artificial intelligence can enhance existing medical tools rather than replace them.

Despite the promising results, researchers caution that more work is needed. The system must be tested in randomized clinical trials to confirm its value in real-world decision making.

They also aim to explore how the model can predict responses to specific treatments. This could further improve its usefulness in guiding care.

Even so, the findings represent meaningful progress. They suggest that faster, more accessible tools may soon help doctors and patients make better decisions.

This research could improve how breast cancer is treated by providing faster and more accurate risk assessments. Patients may receive more personalized care based on their individual risk rather than general guidelines.

The reduced cost and faster results could make advanced testing available to more people. This is especially important in areas where access to genomic testing is limited.

Preserving tissue samples also opens new possibilities. Doctors may be able to perform additional tests later, which could support new treatments or research.

In the long term, this approach could extend beyond breast cancer. Similar systems may be developed for other diseases, helping to improve diagnosis and treatment across medicine.

By combining speed, accuracy, and accessibility, the technology offers a path toward more informed and effective care.

Research findings are available online in the journal Nature Communications.

The original story "New AI tool predicts breast cancer recurrence faster" is published in The Brighter Side of News.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.