Scientists in Germany have developed an artificial intelligence system that can help diagnose brain tumours in minutes using standard microscopic tissue samples.

The system, called Hetairos, was developed by experts in Heidelberg and is designed to classify central nervous system tumours with high speed and accuracy. Researchers say it can provide results in about 12 minutes, compared with the average 12-day wait for some current molecular testing methods.

The technology was developed by a team led by Moritz Gerstung of the German Cancer Research Center and Felix Sahm of Heidelberg Medical Faculty and Heidelberg University Hospital. The research was published in Nature Cancer.

Brain tumour diagnosis has become increasingly complex because many tumours can only be identified accurately by examining both their microscopic appearance and molecular features. These tests often require specialised equipment, advanced laboratory facilities and several days or weeks before results are available.

Hetairos uses digitised images of routinely prepared and stained tissue sections to predict the molecular subtype of a brain or spinal cord tumour. Researchers said the system can identify more than 100 molecular subtypes of central nervous system tumours.

The AI model was trained on more than 11,000 tissue samples from 9,606 patients across 11 medical centres, according to reports on the study. It was designed to support doctors by narrowing possible diagnoses and guiding further testing, rather than replacing molecular diagnostics entirely.

In a comparison with five experienced neuropathologists, Hetairos performed better when both the AI and specialists relied only on standard tissue sections. Medical Xpress reported that the system achieved 68 percent accuracy for its main diagnosis, compared with an average of 30 percent among specialists in the test group.

Researchers said the system could be especially useful where molecular testing is unavailable, delayed or limited by small tissue samples. It may also help doctors make faster decisions about treatment planning.

Moritz Gerstung said the work shows how AI-supported digital pathology can improve medical diagnosis by making complex classification tools faster and more widely accessible.

Experts stressed that Hetairos is a decision-support tool and further clinical use will depend on validation, integration into hospital workflows and regulatory approval.

Also Read : Joint Pain Supplement Glucosamine Linked to Faster Dementia Progression, Study Finds