Data Science Investment Counter —Funding raised by UK data science companies in 2018.
£ 5.640 Million

AI helps to better assess treatment response of brain tumours

Scientists from Heidelberg University Hospital and the German Cancer Research Centre have developed a new machine learning method for the automated image analysis of brain tumours.

According to the authors who worked on the research, machine learning methods trained on standard magnetic resonance imaging (MRI) are more reliable and precise than traditional radiological methods in the treatment of brain tumours.

Gliomas are the most common and most malignant brain tumours in adults. This kind of tumours cannot be entirely removed by surgery, but its spreading can be limited by using chemotherapy or radiotherapy.

Because of this, automated image analysis via machine learning could make a valuable contribution to the individualised treatment of tumours, allowing patients to live longer and better.

One of the main criteria for the assessment of the efficacy of a new therapy for brain tumours is the growth dynamic, determined by MRI. However, the manual measurement of tumour expansion in the contrast-enhanced MRI scans is susceptible to human error and leads to slightly different results.

“This can have a negative effect on the assessment of therapy response and hence the reproducibility and precision of scientific statements based on imaging,” explains Martin Bendszus, Medical Director of the Department of Neuroradiology at the University Hospital in Heidelberg.

Using a database with MRI scans of almost 500 brain tumour patients at Heidelberg University Hospital, the new algorithms were able to automatically locate and recognise brain tumours using artificial neural networks. In addition, the algorithms were also trained to volumetrically measure the individual areas and to precisely assess the response to therapy.

The results were measured and validated in cooperation with the European Organization for Research and Treatment of Cancer.

“The evaluation of more than 2,000 MRI scans of 534 glioblastoma patients from all over Europe shows that our computer-based approach allows a more reliable assessment of therapy response than the conventional method of manual measurement,” said Philipp Kickingereder, leader of the team who conducted the study and researcher at the Department of Neuroradiology at Heidelberg University Hospital.

“We were able to improve the reliability of the assessment by 36 per cent. This can be crucial for the image-based assessment of therapy efficacy in clinical trials. The prediction of overall survival was also more precise with our new method”, he added.

The researchers also designed and evaluated a software infrastructure that enables the complete integration of the new technique into existing radiological infrastructure, to facilitate its spreading across the healthcare sector.

“In this way, we are creating the prerequisites for broad application and fully automated processing and analysis of MRI scans of brain tumours within a few minutes,” explains Klaus Maier-Hein from the German Cancer Research Centre.

“In the future,” he added, “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 such as brain metastases or multiple sclerosis.”

Image via Wikipedia/Steven Keating

SHARE THIS ARTICLE:
BY SHACK15

Co-working space and blog dedicated to all things data science.

Subscribe to our newsletter