Researchers have developed an artificial intelligence-based (AI) method for analysis of brain tumours, paving the way for individualised treatment of tumours. According to the study, published in the The Lancet Oncology, AI machine learning methods, carefully trained on standard magnetic resonance imaging (MRI), are more reliable and precise than established radiological methods in the treatment of gliomas.

Glioma, a type of tumour that occurs in the brain and spinal cord, is common and most malignant of brain tumours in adults. “With this study, we were able to demonstrate the potential of artificial neural networks in radiological diagnostics,” said Philipp Kickingereder from the Heidelberg University in Germany.

For the study, the team included 500 brain tumour patients. Using a reference database with MRI scans of patients, the algorithms automatically recognised and localised brain tumours using artificial neural networks. The algorithms were also enabled to volumetrically measure the individual areas (contrast medium-absorbing tumour portion, peritumoral edema).

“We want to advance the technology for automated high-throughput analysis of medical image data and transfer it not only to brain tumours but also to other diseases like brain metastases or multiple sclerosis,” said Klaus Maier Hein at the varsity.

Glioma tumours often cannot be completely removed by surgery. Chemotherapy or radiotherapy are only effective to a limited extent because tumours are highly resistant. Therefore, new and precisely validated treatment approaches are urgently needed, the team noted.

Researchers have developed an artificial intelligence-based (AI) method for analysis of brain tumours, paving the way for individualised treatment of tumours. According to the study, published in the The Lancet Oncology, AI machine learning methods, carefully trained on standard magnetic resonance imaging (MRI), are more reliable and precise than established radiological methods in the treatment of gliomas.

Glioma, a type of tumour that occurs in the brain and spinal cord, is common and most malignant of brain tumours in adults. “With this study, we were able to demonstrate the potential of artificial neural networks in radiological diagnostics,” said Philipp Kickingereder from the Heidelberg University in Germany.

For the study, the team included 500 brain tumour patients. Using a reference database with MRI scans of patients, the algorithms automatically recognised and localised brain tumours using artificial neural networks. The algorithms were also enabled to volumetrically measure the individual areas (contrast medium-absorbing tumour portion, peritumoral edema).

“We want to advance the technology for automated high-throughput analysis of medical image data and transfer it not only to brain tumours but also to other diseases like brain metastases or multiple sclerosis,” said Klaus Maier Hein at the varsity.

Glioma tumours often cannot be completely removed by surgery. Chemotherapy or radiotherapy are only effective to a limited extent because tumours are highly resistant. Therefore, new and precisely validated treatment approaches are urgently needed, the team noted.