King's College London

Research portal

Machine Learning and Glioma Imaging Biomarkers

Research output: Contribution to journalArticle

Original languageEnglish
Pages (from-to)20-32
Number of pages13
JournalClinical Radiology
Issue number1
Early online date29 Jul 2019
Publication statusPublished - 1 Jan 2020


King's Authors


AIM: To review how machine learning (ML) is applied to imaging biomarkers in neuro-oncology, in particular for diagnosis, prognosis, and treatment response monitoring.MATERIALS AND METHODS: The PubMed and MEDLINE databases were searched for articlespublished before September 2018 using relevant search terms. The search strategy focused onarticles applying ML to high-grade glioma biomarkers for treatment response monitoring,prognosis, and prediction.RESULTS: Magnetic resonance imaging (MRI) is typically used throughout the patientpathway because routine structural imaging provides detailed anatomical and pathologicalinformation and advanced techniques provide additional physiological detail. Using carefullychosen image features, ML is frequently used to allow accurate classification in a variety ofscenarios. Rather than being chosen by human selection, ML also enables image features to beidentified by an algorithm. Much research is applied to determining molecular profiles, his-tological tumour grade, and prognosis using MRI images acquired at the time that patientsfirstpresent with a brain tumour. Differentiating a treatment response from a post-treatment-related effect using imaging is clinically important and also an area of active study(described here in one of two Special Issue publications dedicated to the application of ML inglioma imaging).CONCLUSION: Although pioneering, most of the evidence is of a low level, having beenobtained retrospectively and in single centres. Studies applying ML to build neuro-oncologymonitoring biomarker models have yet to show an overall advantage over those using tradi-tional statistical methods. Development and validation of ML models applied to neuro-oncology require large, well-annotated datasets, and therefore multidisciplinary and multi-centre collaborations are necessary.

Download statistics

No data available

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454