A computer tool tries to read a glioma marker from MRI scans
Original title: MRI-derived habitat radiomics integrated with transformer deep learning for preoperative prediction of high CD9 expression: a novel prognostic marker in adult-type diffuse gliomas.
How far along is this research?
- Lab cells
- Animals
- Review
- Tested in people
This was tested in people. That is the most reliable kind of research we share.
It was built from scans and records already on file, and still needs testing in new patients.
The short version
Researchers built a computer tool that looks at MRI scans to guess whether a brain tumor carries a lot of a protein called CD9.
What was studied. Researchers used stored scans and tumor data from 262 adults with diffuse glioma: A glioma whose cells spread out into the brain around it instead of staying in one lump. That is why the edge on a scan is not a clean line. See the glossary, a type of brain tumor. They trained computer programs on the MRI scans to predict which tumors had high levels of a protein called CD9.
What they found. Combining the different computer programs into one tool worked better than any single program. The tool did very well on the scans it learned from. It did less well on new scans it had not seen before. Patients the tool placed in the high risk group lived for less time than those in the low risk group.
What this means, and what it doesn't
What it could mean: If this holds up, a routine scan might one day give doctors a hint about how a tumor will behave. That could help with planning care. Right now it is a research tool and not something you can ask for.
What it doesn't mean: This is not a treatment and it is not a cure. The tool was built by looking back at scans and records that were already collected. The authors say it still needs to be tested in new patients at more than one hospital. It does not change the care you get today.
Source: PubMed, August 17, 2026 · Read the original
This plain-language summary was written by AI and published automatically after passing our automatic safety checks. How we write.