Search

Brain tumours are the most common solid cancers in children in the UK and are the most common cause of cancer deaths in this age group. Despite current advances in MRI, non-invasive diagnosis of paediatric brain tumours has yet to find its way into routine clinical practice. Radiomics, the high-throughput extraction and analysis of quantitative image features (e.g. texture), offers potential solutions for tumour characterization and decision support. In the search for diagnostic oncological markers, the primary aim of this work was to study the application of MRI texture analysis (TA) for the classification of paediatric brain tumours. A multicentre study was carried out, within a supervised classification framework, on clinical MR images, and a support vector machine (SVM) was trained with 3D textural attributes obtained from conventional MRI. To determine the cross-centre transferability of TA, an assessment of how SVM performs on unseen datasets was carried out through rigorous pairwise testing. The study also investigated the nature of features that are most likely to train classifiers that can generalize well with the data. Finally, the issue of class imbalance, which arises due to some tumour types being more common than others, was explored. For each of the tests carried out through pairwise testing, the optimal area under the receiver operating characteristic curve ranged between 76% and 86%, suggesting that the model was able to capture transferable tumour information. Feature selection results suggest that similar aspects of tumour texture are enhanced by MR images obtained at different hospitals. Our results also suggest that the availability of equally represented classes has enabled SVM to better characterize the data points. The findings of the study presented here support the use of 3D TA on conventional MR images to aid diagnostic classification of paediatric brain tumours., (Copyright © 2017 John Wiley & Sons, Ltd.)

Brain tumours are the most frequently occuring solid tumours affecting childhood, representing 27% of all cancers. The most common posterior fossa tumours are medulloblastoma, pilocytic astrocytoma and ependymoma. Texture Analysis (TA) of Magnetic Resonance Imaging (MRI) aims to represent pixel distributions, intensities and dependencies using mathematically defined features. Such features could potentially provide quantifiable information that is beyond the human vision capabilities, and hence be used to supplement qualitative assessments conducted by radiologists. The primary aim of this study was to carry out a multicentre investigation on the efficacy of 3D TA for diagnostic classification of childhood brain tumours, using conventional MRI images. The data used had been acquired at three different hospitals and consisted of pre-contrast T1 and T2-weighted MRI series, obtained from 121 children diagnosed with medulloblastoma, pilocytic astrocytoma and ependymoma. Using 3D textural features, based on first, second and higher order statistical methods, a support vector machine (SVM) classifier was trained and tested using the leave-one-out cross-validation (LOOCV) approach. An essential outcome of this study is that 3D TA demonstrated a good overall performance, when used on data acquired from a number of centres and using scanners made by different manufacturers and at different magnetic field strengths.

Brain and central nervous system (CNS) tumours form the second most common group of cancers in children in the UK, accounting for 27% of all childhood cancers. Initial assessment of tumours from MRI scans is usually performed qualitatively, via radiologists' visual inspection. However, different brain tumours do not always demonstrate clear differences in physical appearance, so a diagnosis is usually made via histopathological examination of biopsy samples taken through surgery. This gives rise to the need for accurate, yet non-invasive diagnostic aids. In a previous study, we demonstrated the potential of MRI texture analysis in capturing quantitative information about paediatric brain tumours. In this work, we carry out a preliminary investigation on the use of 3D (volumetric) texture analysis of T1 and T2-weighted MR images in order to classify paediatric brain tumours. We then compare its performance with the traditional 2D texture analysis approach. Our preliminary findings are very encouraging and show that 3D textural features are capable of capturing more discriminative information about the tumours than the traditional 2D approach. However, it remains necessary to expand the work further to include larger cohorts and additional modalities.

Despite current advances in neuro-imaging, the characterisation of paediatric brain tumours and neurological disorders is very challenging. Whilst Magnetic Resonance Imaging (MRI) provides images of superb clarity, it gives little information on how aggressive a tumour is. It is also very difficult to visually inspect any underlying textural patterns between tumours on MR images. This gives rise to the need for a quantitative means of analysing MR images for characterising tumours. In this work, we present a preliminary investigation into the effectiveness of texture analysis as a quantitative approach for classifying paediatric brain tumours.