Semi-automated brain tumor segmentation on multi-parametric MRI using regularized non-negative matrix factorization

Abstract

Background: Segmentation of gliomas in multi-parametric (MP-)MR images is challenging due to their heterogeneous nature in terms of size, appearance and location. Manual tumor segmentation is a time-consuming task and clinical practice would benefit from (semi-) automated segmentation of the different tumor compartments.

Methods: We present a semi-automated framework for brain tumor segmentation based on non-negative matrix factorization (NMF) that does not require prior training of the method. L1-regularization is incorporated into the NMF objective function to promote spatial consistency and sparseness of the tissue abundance maps. The pathological sources are initialized through user-defined voxel selection. Knowledge about the spatial location of the selected voxels is combined with tissue adjacency constraints in a post-processing step to enhance segmentation quality. The method is applied to an MP-MRI dataset of 21 high-grade glioma patients, including conventional, perfusion-weighted and diffusion-weighted MRI. To assess the effect of using MP-MRI data and the L1-regularization term, analyses are also run using only conventional MRI and without L1-regularization. Robustness against user input variability is verified by considering the statistical distribution of the segmentation results when repeatedly analyzing each patient's dataset with a different set of random seeding points.

Results: Using L1-regularized semi-automated NMF segmentation, mean Dice-scores of 65%, 74 and 80% are found for active tumor, the tumor core and the whole tumor region. Mean Hausdorff distances of 6.1 mm, 7.4 mm and 8.2 mm are found for active tumor, the tumor core and the whole tumor region. Lower Dice-scores and higher Hausdorff distances are found without L1-regularization and when only considering conventional MRI data.

Conclusions: Based on the mean Dice-scores and Hausdorff distances, segmentation results are competitive with state-of-the-art in literature. Robust results were found for most patients, although careful voxel selection is mandatory to avoid sub-optimal segmentation.

Keywords: Brain tumors; MRI; Non-negative matrix factorization; Segmentation; Unsupervised classification.

Fig. 1

Illustration of morphological post-processing after initial semi-automated NMF based segmentation of necrosis a and active tumor c. Step 1: false positives are removed by withholding only the connected components closest to the user-defined seeding points (marked by cursor arrows) for necrosis (b) and for active tumor (d). Step 2: spatial adjacency of the connected components in the preliminary necrosis mask (green) to the withheld active tumor mask (red) is verified in (e). The final necrosis mask is shown in yellow in (f)

Fig. 2

First row: coregistered MP-MRI maps of a GBM patient, left to right: T1C, FLAIR, rCBV, ADC. Second row, left to right: NMF abundance maps for active tumor, necrosis and edema. The final segmentation masks are shown on the right for active tumor (red), necrosis (yellow) and edema (blue)

Fig. 3

Comparison of the segmentation results of the pathological tissue regions obtained for a GBM patient using the different NMF methodologies. The left figure of each row shows the obtained segmentation for active tumor (purple), necrosis (red) and edema (yellow). The second to fourth figure of each row show the individual segmentation for active tumor, necrosis and edema, respectively. The top row shows manual segmentation, whereas the other rows show the overlap between the NMF (blue) and manual (green) segmentation result. Segmentation overlap is marked in cyan

Fig. 4

Boxplots showing the dispersion of the Dice-scores for active tumor. Boxplots show quartile ranges of the Dice-scores, ’+’ indicates outliers

Fig. 5

Boxplots showing the dispersion of the Dice-scores for the tumor core. Boxplots show quartile ranges of the Dice-scores, ’+’ indicates outliers

Fig. 6

Boxplots showing the dispersion of the Dice-scores for the whole tumor region. Boxplots show quartile ranges of the Dice-scores, ‘+’ indicates outliers

Fig. 7

Example of a bad segmentation result due to suboptimal voxel selection. A close-up of a GBM lesion on an axial T1C slice (a). The manually segmented necrotic region is shown in yellow (b), the selected necrotic voxel is marked in red. Segmentation of the active tumor region (c) and necrotic region (d) on several slices, blue indicates NMF segmentation, green indicates manual segmentation and cyan indicates overlap