CT Scans Predicted Lung Cancer Survival Outcomes

A new AI-based imaging score may help identify which non-small cell lung cancer patients have a better prognosis.

Updated on Sept. 21, 2026 in Cancer

Isometric editorial illustration of a complex branching vascular lattice structure, symbolizing medical data analysis.
Researchers at Highwise Health developed a deep learning model that utilizes tumor vascular patterns on CT scans to more accurately predict survival outcomes in lung cancer patients. AI Illustration. Upload story photo >

Live Poll

Do you believe noninvasive imaging should be prioritized to help guide cancer immunotherapy treatment?

Researchers have developed a deep learning tool that analyzes tumor blood vessel patterns on CT scans to gauge patient outcomes. In a recent study of 321 patients, those with lower vascular risk scores demonstrated improved progression-free and overall survival.

Why it matters

Understanding tumor vascular health is vital because morphological abnormalities often contribute to treatment resistance in lung cancer. This new metric may help clinicians better predict how patients respond to immune checkpoint inhibitors.

This validation study evaluated 321 patients with non-small cell lung cancer (NSCLC) using a deep learning imaging metric. Findings showed that lower vascular risk scores were associated with longer progression-free and overall survival compared to higher risk scores.

The details

The model works by learning complex representations of vascular morphology from standard medical imaging. It utilizes Gaussian mixture modeling to quantify the degree to which a tumor's blood vessel structure deviates from normal, healthy vascular anatomy. When combined with existing markers like PD-L1 expression, this tool provides additional discrimination for patient outcomes.

Timeline

  1. September 21, 2026: The research findings were published.

Health Landscape

This development represents a shift toward using artificial intelligence to decode tumor microenvironments beyond traditional biopsy methods. It builds on the broader effort to overcome treatment resistance in non-small cell lung cancer by identifying prognostic imaging biomarkers.

If you or a loved one are undergoing care for lung cancer, this research highlights the growing role of advanced imaging analysis in treatment planning. You may want to speak with your oncologist about how specific imaging patterns might inform your prognosis or treatment trajectory.

The takeaway

Tumor vascular patterns identified on standard scans offer a new way to understand potential cancer treatment responses. Patients should continue to discuss all imaging results and prognosis assessments directly with their clinical care team.

Further reading

For more on evolving diagnostic tools, visit our Cancer section.

More information

View the complete peer-reviewed research article for full technical details.

Live Poll

Do you believe noninvasive imaging should be prioritized to help guide cancer immunotherapy treatment?