Texture analysis- and support vector machine-assisted diffusional kurtosis imaging may allow in vivo gliomas grading and IDH-mutation status prediction: a preliminary study

Abstract

We sought to investigate, whether texture analysis of diffusional kurtosis imaging (DKI) enhanced by support vector machine (SVM) analysis may provide biomarkers for gliomas staging and detection of the IDH mutation. First-order statistics and texture feature extraction were performed in 37 patients on both conventional (FLAIR) and mean diffusional kurtosis (MDK) images and recursive feature elimination (RFE) methodology based on SVM was employed to select the most discriminative diagnostic biomarkers. The first-order statistics demonstrated significantly lower MDK values in the IDH-mutant tumors. This resulted in 81.1% accuracy (sensitivity = 0.96, specificity = 0.45, AUC 0.59) for IDH mutation diagnosis. There were non-significant differences in average MDK and skewness among the different tumour grades. When texture analysis and SVM were utilized, the grading accuracy achieved by DKI biomarkers was 78.1% (sensitivity 0.77, specificity 0.79, AUC 0.79); the prediction accuracy for IDH mutation reached 83.8% (sensitivity 0.96, specificity 0.55, AUC 0.87). For the IDH mutation task, DKI outperformed significantly the FLAIR imaging. When using selected biomarkers after RFE, the prediction accuracy achieved 83.8% (sensitivity 0.92, specificity 0.64, AUC 0.88). These findings demonstrate the superiority of DKI enhanced by texture analysis and SVM, compared to conventional imaging, for gliomas staging and prediction of IDH mutational status.

Machine learning assisted DSC-MRI radiomics as a tool for glioma classification by grade and mutation status

“…By contrast to first-order statistics, second-and higher-order statistical analyses, which allow measures of not only local voxel-wise values but also incorporate neighbouring information, have been suggested as more reliable and elaborative methods for characterisation of the region-of-interest [13]. Texture analysis, as a contextual quantification method, has already embarked in the medical imaging literature as a method that can detect tissue heterogeneity and complexity [13,14]. Radiomics, applied in the clinical context of glioblastoma, have generally been performed on anatomical sequences and applied to distinguish between GBM subtypes [15], for prediction of survival rates [16] and prognosis [17], prediction of response to treatment [18], as well as risk stratification [19].…”

MRI Radiomic Features to Predict IDH1 Mutation Status in Gliomas: A Machine Learning Approach using Gradient Tree Boosting

“…Recent improvements in ML algorithms and computational power provide an attractive venue for exploring MR radiomic features, an excellent fit for ML-approach analysis that considers the large data size and multimodal nature. Therefore, ML methods have been recently explored to predict glioma genetic biomarkers from MRI radiomic features [10][11][12][21][22][23][24][25][26][27][28][29][30][31][32][33].…”

“…Recent investigations on the use of ML and MRI-radiomics to predict IDH1 genotype have primarily explored the SVM [21,25,30] and RF [10,22,24,[27][28][29] models. Some studies only used conventional MRI sequences and achieved AUC ranging 0.84-0.96 [10,[22][23][24][25]30].…”

“…Some studies only used conventional MRI sequences and achieved AUC ranging 0.84-0.96 [10,[22][23][24][25]30]. Others explored the added value of advanced MRI imaging, such as MR diffusion or perfusion, with mixed results [21,[26][27][28][29]31]. The highest performance of predicting IDH1 genotype with AUC of 0.96 was observed with a RF model that was trained with conventional MRI, but this study focused only on patients with glioblastomas [10].…”

Radiogenomics: bridging imaging and genomics

“…Isocitrate dehydrogenase-1 (IDH) mutation is commonly used in the clinic to stratify patients, often in conjunction with other co-mutations [49] and it received the most attention in brain radiogenomic research. IDH-mutated gliomas occurred most frequently in the rostral extension of the lateral ventricles of the frontal lobe [50] and were linked to tumor size [51], local pattern of intensities [52], PET features [53,54], angular standard deviation (tumor boundary irregularity) [41], mean diffusional kurtosis [55], and apparent diffusion coefficient (ADC) [56] as well as part of "radiomic signatures" in artificial intelligence models [20,[57][58][59][60][61][62]. Similarly, 1p/19q co-deletion [62][63][64][65], a widely used prognostic biomarker for brain tumors [66], and EGFR mutation [67][68][69][70] were thoroughly covered by different teams, having been linked to a wide array of MRI features.…”