Diffusion-weighted MRI radiomics of spine bone tumors: feature stability and machine learning-based classification performance

Abstract

Purpose: To evaluate stability and machine learning-based classification performance of radiomic features of spine bone tumors using diffusion- and T2-weighted magnetic resonance imaging (MRI).

Material and methods: This retrospective study included 101 patients with histology-proven spine bone tumor (22 benign; 38 primary malignant; 41 metastatic). All tumor volumes were manually segmented on morphologic T2-weighted sequences. The same region of interest (ROI) was used to perform radiomic analysis on ADC map. A total of 1702 radiomic features was considered. Feature stability was assessed through small geometrical transformations of the ROIs mimicking multiple manual delineations. Intraclass correlation coefficient (ICC) quantified feature stability. Feature selection consisted of stability-based (ICC > 0.75) and significance-based selections (ranking features by decreasing Mann-Whitney p-value). Class balancing was performed to oversample the minority (i.e., benign) class. Selected features were used to train and test a support vector machine (SVM) to discriminate benign from malignant spine tumors using tenfold cross-validation.

Results: A total of 76.4% radiomic features were stable. The quality metrics for the SVM were evaluated as a function of the number of selected features. The radiomic model with the best performance and the lowest number of features for classifying tumor types included 8 features. The metrics were 78% sensitivity, 68% specificity, 76% accuracy and AUC 0.78.

Conclusion: SVM classifiers based on radiomic features extracted from T2- and diffusion-weighted imaging with ADC map are promising for classification of spine bone tumors. Radiomic features of spine bone tumors show good reproducibility rates.

Keywords: Machine learning; Radiomics; Reproducibility; Spine; Texture analysis; Tumor.

Fig. 1

Example of tumor segmented by the radiologist. The tumor is segmented on the T2-weighted image (a) but the same segmentation has been used as a mask to extract the radiomic features from the apparent diffusion coefficient map (b) as well

Fig. 2

Translated versions of the same region of interest used for the stability analysis. (a) Upward translation. (b) Downward translation. (c) Translation to the right. (d) Translation to the left

Fig. 3

Workflow of the radiomic classifier training

Fig. 4

Results of the best radiomic classifiers. (a) Confusion matrix on which sensitivity, specificity and accuracy have been computed. (b) Receiver operating characteristic (ROC) curve, with the black dot representing the actual sensitivity and specificity of the classifier

Fig. 5

Boxplots showing the distribution of the values of mean apparent diffusion coefficient (ADC). (a) Values of this study. (b) Distributions observed in [5]