Technology meets health: soon, AI could offer treatment for a brain tumor



[ad_1]

LONDON: Researchers have developed a method of badyzing brain tumors based on artificial intelligence (AI), paving the way for individualized treatment of tumors.

According to the study published in The Lancet Oncology, AI's automatic learning methods, carefully trained in standard magnetic resonance imaging (MRI), are more reliable and more accurate than radiological methods established in the treatment of glioma.

Gliomas, a type of tumor that occurs in the brain and spinal cord, are the most common brain tumors and the most malignant in adults.

"Through this study, we were able to demonstrate the potential of artificial neural networks in radiological diagnosis," said Philipp Kickingereder of the University of Heidelberg in Germany.

For the study, the team included 500 patients with brain tumors. Using a reference database with MRI scans of patients, the algorithms automatically recognized and localized brain tumors using artificial neural networks.

The algorithms have also been used to volumetrically measure the individual zones (tumor portion absorbing the contrast medium, peritumoral edema).

"We want to advance the technology of automated high-throughput badysis of medical image data and transfer it not only to brain tumors, but also to other diseases such as brain metastases or sclerosis." in patches, "said Klaus Maier Hein at the university.

Glioma tumors often can not be completely removed by surgery. Chemotherapy or radiotherapy are only effective to a limited extent because the tumors are very resistant. As a result, new and precisely validated therapeutic approaches are urgently needed, the team noted.

Next Horizon of Science: AI to diagnose cerebral hemorrhage, spaceship for tra …

Tech Surprises

Sep 7, 2018

From the AI ​​who can diagnose a stroke to scientists who use antimatter for a truck ride, here is all that is exciting for the technologies that will shape our future. (Text: Rajarshi Bhattacharjee)

[ad_2]
Source link