Testing AI on brain scans taken before treatment starts
Original title: Deep learning prediction of overall survival in glioblastoma patients using pre-chemoradiotherapy MRI and clinical data.
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.
This was a computer program tested on old scans, not a treatment anyone can get.
The short version
Researchers tested a computer program that reads brain scans before treatment begins.
What was studied. Researchers used 1,575 brain scans from 175 people treated for 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. The scans were taken between 2018 and 2023, all before radiation and chemo started.
What they found. Using scans alone, the program sorted short and long survivors correctly 81.08% of the time. A version that sorted people into three groups was correct 87.18% of the time. Adding health details from the patients' records did not work better than the scans alone.
What this means, and what it doesn't
What it could mean: Someday doctors may use an early scan to see who faces a harder road. That could help them plan care sooner. For now this is only a research tool.
What it doesn't mean: This is not a new treatment. The program only tried to guess what would happen. It did not change anything for the people in the study. It was tested on old scans, not in real clinics with new patients. It is not a cure, and it says nothing certain about any one person.
Source: PubMed, October 13, 2026 · Read the original
This plain-language summary was written by AI and published automatically after passing our automatic safety checks. How we write.