1
|
Distinguishing between paediatric brain tumour types using multi-parametric magnetic resonance imaging and machine learning: A multi-site study. NEUROIMAGE-CLINICAL 2020; 25:102172. [PMID: 32032817 PMCID: PMC7005468 DOI: 10.1016/j.nicl.2020.102172] [Citation(s) in RCA: 18] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 10/22/2019] [Revised: 12/04/2019] [Accepted: 01/10/2020] [Indexed: 12/12/2022]
Abstract
The imaging and subsequent accurate diagnosis of paediatric brain tumours presents a radiological challenge, with magnetic resonance imaging playing a key role in providing tumour specific imaging information. Diffusion weighted and perfusion imaging are commonly used to aid the non-invasive diagnosis of children's brain tumours, but are usually evaluated by expert qualitative review. Quantitative studies are mainly single centre and single modality. The aim of this work was to combine multi-centre diffusion and perfusion imaging, with machine learning, to develop machine learning based classifiers to discriminate between three common paediatric tumour types. The results show that diffusion and perfusion weighted imaging of both the tumour and whole brain provide significant features which differ between tumour types, and that combining these features gives the optimal machine learning classifier with >80% predictive precision. This work represents a step forward to aid in the non-invasive diagnosis of paediatric brain tumours, using advanced clinical imaging.
Collapse
|
2
|
Berthet C, Lei H, Gruetter R, Hirt L. Early Predictive Biomarkers for Lesion After Transient Cerebral Ischemia. Stroke 2011; 42:799-805. [DOI: 10.1161/strokeaha.110.603647] [Citation(s) in RCA: 33] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Background and Purpose—
Despite the improving imaging techniques, it remains challenging to predict the outcome early after transient cerebral ischemia. The aim of this study was thus to identify early metabolic biomarkers for outcome prediction.
Methods—
We modeled transient ischemic attacks and strokes in mice. Using high-field MR spectroscopy, we correlated early changes in the neurochemical profile of the ischemic striatum with histopathologic alterations at a later time point.
Results—
A significant increase in glutamine was measured between 3 hours and 8 hours after all ischemic events followed by reperfusion independently of the outcome and can thus be considered as an indicator of recent transient ischemia. On the other hand, a reduction of the score obtained by summing the concentrations of N-acetyl aspartate, glutamate, and taurine was a good predictor of an irreversible lesion as early as 3 hours after ischemia.
Conclusions—
We identified biomarkers of reversible and irreversible ischemic damage, which can be used in an early predictive evaluation of stroke outcome.
Collapse
Affiliation(s)
- Carole Berthet
- From the Department of Clinical Neurosciences (C.B., L.H.), Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland; the Laboratory of Functional and Metabolic Imaging (H.L., R.G.), Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland; the Department of Radiology (H.L., R.G.), University of Lausanne, Lausanne, Switzerland; and the Department of Radiology (R.G.), University of Geneva, Geneva, Switzerland
| | - Hongxia Lei
- From the Department of Clinical Neurosciences (C.B., L.H.), Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland; the Laboratory of Functional and Metabolic Imaging (H.L., R.G.), Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland; the Department of Radiology (H.L., R.G.), University of Lausanne, Lausanne, Switzerland; and the Department of Radiology (R.G.), University of Geneva, Geneva, Switzerland
| | - Rolf Gruetter
- From the Department of Clinical Neurosciences (C.B., L.H.), Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland; the Laboratory of Functional and Metabolic Imaging (H.L., R.G.), Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland; the Department of Radiology (H.L., R.G.), University of Lausanne, Lausanne, Switzerland; and the Department of Radiology (R.G.), University of Geneva, Geneva, Switzerland
| | - Lorenz Hirt
- From the Department of Clinical Neurosciences (C.B., L.H.), Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland; the Laboratory of Functional and Metabolic Imaging (H.L., R.G.), Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland; the Department of Radiology (H.L., R.G.), University of Lausanne, Lausanne, Switzerland; and the Department of Radiology (R.G.), University of Geneva, Geneva, Switzerland
| |
Collapse
|