Distinguishing between paediatric brain tumour types using multi-parametric magnetic resonance imaging and machine learning: A multi-site study

Highlights • Individual diffusion and perfusion imaging radiomics can provide limited information to distinguish between tumour types.• Combining functional imaging with machine learning, it is possible to distinguish between the three main tumour types with high accuracy.• Whole brain features further contribute to classifier accuracy.• Target feature selection outperforms non-biased methods.

[1]  Karl-Josef Langen,et al.  Imaging of sodium in the brain: a brief review , 2016, NMR in biomedicine.

[2]  A. Panigrahy,et al.  Apparent diffusion coefficient histogram metrics correlate with survival in diffuse intrinsic pontine glioma: a report from the Pediatric Brain Tumor Consortium. , 2016, Neuro-oncology.

[3]  Theodoros N. Arvanitis,et al.  University of Birmingham Application of pattern recognition techniques for classification of paediatric brain tumours by in vivo 3T 1H MRS – A multi centre study , 2017 .

[4]  Theodoros N. Arvanitis,et al.  (1)H magnetic resonance spectroscopy in the diagnosis of paediatric low grade brain tumours. , 2013, European journal of radiology.

[5]  R. Faghihi,et al.  Magnetic Resonance Spectroscopy and its Clinical Applications: A Review. , 2017, Journal of medical imaging and radiation sciences.

[6]  N. Girard,et al.  Cerebral tumors: specific features in children. , 2014, Diagnostic and interventional imaging.

[7]  D. Auer,et al.  Multi-centre reproducibility of diffusion MRI parameters for clinical sequences in the brain , 2015, NMR in biomedicine.

[8]  Nitesh V. Chawla,et al.  SMOTE: Synthetic Minority Over-sampling Technique , 2002, J. Artif. Intell. Res..

[9]  C. Trevenen,et al.  Middle cerebral artery dissection. , 1996, Neuropediatrics.

[10]  Theodoros N. Arvanitis,et al.  Diagnosing relapse in children's brain tumors using metabolite profiles. , 2014, Neuro-oncology.

[11]  Pascal J. Kieslich,et al.  Gadolinium retention in the dentate nucleus and globus pallidus is dependent on the class of contrast agent. , 2015, Radiology.

[12]  B. Erickson,et al.  Machine Learning for Medical Imaging. , 2017, Radiographics : a review publication of the Radiological Society of North America, Inc.

[13]  Denis Le Bihan,et al.  Looking into the functional architecture of the brain with diffusion MRI , 2003, Nature Reviews Neuroscience.

[14]  N. Rangwala.  Reduction of Artifacts Arising From Non-Ideal Gradients in Fast Magnetic Resonance Imaging , 2011 .

[15]  James M. Provenzale,et al.  Principles of T2*‐weighted dynamic susceptibility contrast MRI technique in brain tumor imaging , 2015, Journal of magnetic resonance imaging : JMRI.

[16]  Theodoros N. Arvanitis,et al.  Texture analysis of T1- and T2-weighted MR images and use of probabilistic neural network to discriminate posterior fossa tumours in children , 2014, NMR in biomedicine.

[17]  A. Peet,et al.  A comparison of pseudo-continuous arterial spin labelling and dynamic susceptibility contrast MRI with and without contrast agent leakage correction in paediatric brain tumours. , 2019, The British journal of radiology.

[18]  C. Clark,et al.  Discrimination of paediatric brain tumours using apparent diffusion coefficient histograms , 2012, European Radiology.

[19]  Y. Ra,et al.  Advanced MRI for Pediatric Brain Tumors with Emphasis on Clinical Benefits , 2017, Korean journal of radiology.

[20]  C. Clark,et al.  Arterial spin labelling and diffusion-weighted imaging in paediatric brain tumours , 2019, NeuroImage: Clinical.

[21]  Theodoros N. Arvanitis,et al.  Application of pattern recognition techniques for classification of pediatric brain tumors by in vivo 3T 1H‐MR spectroscopy—A multi‐center study , 2017, Magnetic resonance in medicine.

[22]  J. Baron,et al.  Clinical review: Imaging in ischaemic stroke – implications for acute management , 2007, Critical care.

[23]  Robert H Christenson,et al.  Stroke biomarkers: progress and challenges for diagnosis, prognosis, differentiation, and treatment. , 2010, Clinical chemistry.

[24]  Theodoros N. Arvanitis,et al.  Radiomics in paediatric neuro‐oncology: A multicentre study on MRI texture analysis , 2018, NMR in biomedicine.

[25]  Yuen-Li Chung,et al.  Current Opportunities and Challenges of Magnetic Resonance Spectroscopy, Positron Emission Tomography, and Mass Spectrometry Imaging for Mapping Cancer Metabolism In Vivo , 2014, BioMed research international.

[26]  P. Hluštík,et al.  The role of magnetic resonance imaging for acute ischemic stroke. , 2009, Biomedical papers of the Medical Faculty of the University Palacky, Olomouc, Czechoslovakia.

[27]  Pieter Wesseling,et al.  Accurate classification of childhood brain tumours by in vivo ¹H MRS - a multi-centre study. , 2013, European journal of cancer.