← All research updates

Can computer tools tell real tumor growth from treatment swelling?

Original title: MRI-Based Machine Learning to Differentiate True Progression from Pseudoprogression in Glioblastoma: An Updated Systematic Review and Meta-Analysis.

How far along is this research?

This looks across many earlier studies rather than running a new one.

This is a review of past studies, and the tools are not used in care yet.

The short version

Computer tools that read MRI scans can help spot if a brain tumor is truly growing, but they are not ready for daily care.

What was studied. Researchers pulled together thirty-one studies with 2,344 patients who had glioblastoma: A glioma that is given grade 4. Grade 4 is the highest grade there is. The tumor grows fast. Treatment usually starts soon after it is found. It often means surgery, and then radiation and chemotherapy. See the glossary. Each study used a computer program to read MRI scans and judge if the tumor was really growing.

What they found. Only seven test groups, with 219 patient observations, were used to check the tools on new patients. In those checks the tools were right more often than not, but they still got some cases wrong. The tools looked better on the patients they were built from than on new ones. Many studies also had weak methods or left out key details.

What this means, and what it doesn't

What it could mean: Right now your doctor cannot use one of these tools to make this call. Teams still rely on repeat scans over time, and sometimes surgery, to see what is going on. These tools may help with that choice later.

What it doesn't mean: This is not a new treatment and not a cure. No tool here is approved or part of normal care. The report only looks back at past studies, and the authors say bigger, clearer studies are needed before clinics can use these tools.

Source: PubMed, October 19, 2026 · Read the original

This plain-language summary was written by AI and published automatically after passing our automatic safety checks. How we write.

Report a problem with this summary