The study shows that pictures taken during a routine breast scan can reveal hidden clues about the tumor’s immune environment.  

Radiologists used a technique called radiomics, which turns an image into thousands of tiny data points.  

By feeding these numbers into a computer model, the researchers could predict whether a protein called PD‑L1 was present in the cancer cells.  

PD‑L1 is important because it tells doctors if a tumor might respond to certain immunotherapy drugs.

The team also looked at how dense the breast tissue is, a factor that can make cancers harder to see on scans.  

They found that combining density information with the radiomic data improved the accuracy of their predictions.  

This means a single mammogram could give both a cancer warning and an insight into the best treatment plan.

The research suggests that future imaging could be more than a diagnostic tool; it could become a decision‑making aid.  

If clinicians can know the PD‑L1 status before surgery, they might choose immunotherapy earlier and avoid unnecessary treatments.  

The method is non‑invasive, uses existing equipment, and could be rolled out quickly in many hospitals.

While the results are promising, larger studies are needed to confirm that the predictions hold true across different populations.  

The work also highlights how artificial intelligence can turn ordinary scans into powerful sources of biological information.

Breast Cancer Imaging Helps Spot Immune Signals

Actions