AI Tool Could Speed Brain Tumor Diagnosis from Days to Minutes
AI Tool Could Speed Brain Tumor Diagnosis from Days to Minutes
————————————-
Researchers in Germany have developed an artificial intelligence system that could sharply reduce the time needed to classify brain and spinal cord tumors, potentially helping doctors begin treatment decisions faster.
The system, called Hetairos, was developed by scientists linked to the German Cancer Research Center, Heidelberg University and other medical centers. It uses deep learning to analyze standard stained tissue sections and predict the molecular subtype of tumors, a process that usually requires complex laboratory testing.
In the reported study, Hetairos produced results in about 12 minutes after tissue images were digitized, compared with an average of around 12 days for full molecular diagnostics. The system was trained and tested on more than 11,000 digitized tissue samples from 9,606 patients across 11 medical centers on four continents.
Researchers said Hetairos can distinguish 102 molecular tumor subtypes, covering most of the current World Health Organization classification of central nervous system tumors. In high-confidence cases, its accuracy was reported at about 87 to 88 percent.
Scientists emphasized that the tool is not meant to replace molecular testing, but to support and accelerate diagnosis. It could be especially useful in regions with limited laboratory resources or in cases where there is not enough tumor material for genetic testing.
