Aims and objectives
Gliomas are the most aggressive primary brain tumors,
presenting poor survival rates,
while the accurate preoperative grade classification is of main clinical importance,
related to early prognosis and precise selection of the therapeutic approach.
According to the World Health Organization (WHO) grading system [1],
gliomas are subdivided into four categories considering their malignancy status,
i.e.
grades I,
II (low grade) and grades III,
IV (high grade).
To date,
several studies have reported that MRI may supportively contribute in tumor heterogeneity assessment,
overcoming sample-biopsy limitations,
towards...
Methods and materials
Multiparametric-MRI acquisition and Data post-processing:
Forty patients initially diagnosed with Low- or High-Grade Gliomas (20 LGG & 20 HGG) underwent MRI performed on a 3-Tesla MR whole-body scanner,
applying an advanced imaging examination protocol including,
conventional MRI (T1W-C,
T2W-FSE,
T2W-FLAIR),
MR Spectroscopy (1H-MRS),
Diffusion Tensor Imaging (DTI) and Dynamic Susceptibility Contrast Enhanced MRI (DSCE),
using a 4-channel birdcage and an 8-channel phased-array head coil.
Prior to this retrospective study,
Local Institutional Review Board approval and patient informed consent was obtained (Fig.
1).
FSL software,
was...
Results
The evaluation of different feature subsets with linear ‘SMO’,
has nominated the adaptation of 21 SVM-RFE top ranked features,
shown in Table 1,
which provide the highest discriminating ability between LGGs and HGGs.
Also,
Lipids/Cr metabolic ratio was the highest ranked feature.
As shown in Table 1,
all MRI modalities/parameters have contributed in the final feature set,
except for DTI’s Fractional Anisotropy (FA).
In addition,
8 features where histogram-based and 12 features where textural-based,
while GLCM features were much more statistically significant than GLRLM (11vs1)....
Conclusion
In the present study,
radiomic analysis on a 3T mp-MRI dataset was performed for glioma grade classification between low- and high-grade tumors,
demonstrating 95.5% Accuracy and 95.5% AUC in predicting glioma grades,
utilizing 21 mp-MRI radiomic features.
The justification for implementing the specific SVM feature selection and classification methods,
is based on the predictive robustness indicated by similar studies regarding glioma grading in the past.
In a computer-aided-diagnostic approach Chen et al.
[14] utilized SVM-RFE for selecting textural features,
derived from CNN-based segments of conventional...
Personal information
Corresponding Author:
Dr.
Ioannis Tsougos,
Associate Professor of Medical Physics,Medical School,
University of Thessaly
Visiting Senior Researcher,
Neuroimaging Department,
King’s College London
Tel/Fax: +302413501863
email:
[email protected],
[email protected]
References
Louis DN,
Ohgaki H,
Wiestler OD,
Cavenee WK,
Burger PC,
Jouvet A,
Scheithauer BW,
Kleihues P.,2007.
The 2007 WHO classification of tumours of the central nervous system.,
Acta Neuropathologica 114:97-109
Kevin Li-Chun Hsieh,
Chung-Ming Lo,
Chih-Jou Hsiao,
Computer-aided grading of gliomas based on local and global MRI features,
Computer methods and programs in biomedicine,139(2017):31-38.
Lee J1,
Choi SH,
Kim JH,
Sohn CH,
Lee S,
Jeong J.,
Glioma grading using apparent diffusion coefficient map: application of histogram analysis based on automatic segmentation.,
NMR Biomed.2014Sep;27(9):1046-52.
Svolos,
P.,...