Advanced imaging and AI help classify child brain tumours
Diffusion weighted imaging and machine learning can successfully classify the diagnosis and characteristics of common types of paediatric brain tumours, a UK-based multi-centre study has found. This means that the tumour can be characterised and treated more efficiently.
The biggest cause of death from cancer in children is brain tumours in part of the brain called the posterior fossa. Within this area there are three main types of brain tumour, and being able to characterise them quickly and efficiently can be challenging. Currently a qualitative assessment of MRI by radiologists is used; however, overlapping radiological characteristics can make it difficult to distinguish which type of tumour it is, without the confirmation of biopsy.
A team of UK researchers, led by the University of Birmingham, have now found that tumour diagnostic classification can be improved by using non-invasive diffusion weighted imaging when combined with machine learning (artificial intelligence). Their breakthrough has been published in the journal Scientific Reports.
Diffusion weighted imaging involves the use of specific advanced MRI sequences, as well as software that generates images from the resulting data that uses the diffusion of water molecules to generate contrast in an MRI. One can then extract an apparent diffusion coefficient (ADC) map, analysed values of which can be used to tell you more about the tumour.
The study involved 117 patients from five primary treatment centres across the UK, with scans from 12 different hospitals on a total of 18 different scanners. The images were then analysed and region of interests were drawn by both an experienced radiologist and an expert scientist in paediatric neuroimaging. Values from the analysis of ADC maps from these images’ regions were fed to AI algorithms to successfully and non-invasively distinguish the three most common types of paediatric posterior fossa brain tumours.
“When a child comes to hospital with symptoms that could mean they have a brain tumour, that initial scan is such a difficult time for the family and understandably they want answers as soon as possible,” said study co-author Professor Andrew Peet, from the University of Birmingham and Birmingham Children’s Hospital. “Here we have combined readily available scans with artificial intelligence to provide high levels of diagnostic accuracy that can start to give some answers.
“Previous studies using these techniques have largely been limited to single expert centres. Showing that they can work across such a large number of hospitals opens the door to many children benefiting from rapid non-invasive diagnosis of their brain tumour… we are working hard now to start making these artificial intelligence techniques widely available.”
Study co-author Professor Theo Arvanitis, Director of the Institute of Digital Healthcare at WMG, the University of Warwick, added, “If this advanced imaging technique, combined with AI technology, can be routinely enrolled into hospitals, it means that childhood brain tumours can be characterised and classified more efficiently, and in turn means that treatments can be pursued in a quicker manner with favourable outcomes for children suffering from the disease.”
Researchers have successfully used machine learning to complete cumbersome materials science...
Researchers have found two new mammogram-based measures for breast cancer risk, which could...
Using machine learning, researchers have identified distinct patterns of coordinated activity...