MRI Brain Tumor Segmentation using Random Forests and Fully Convolutional Networks
This work addresses the problem of accurate brain tumor segmentation for medical imaging applications, presenting an incremental improvement over existing methods.
The paper tackles automated brain tumor segmentation in MRI images by combining fully convolutional networks and hand-crafted texton features with random forest classification, achieving mean Dice scores of 0.86, 0.78, and 0.66 for whole tumor, core, and enhancing tumor segmentation on the BRATS 2017 dataset.
In this paper, we propose a novel learning based method for automated segmentation of brain tumor in multimodal MRI images, which incorporates two sets of machine -learned and hand crafted features. Fully convolutional networks (FCN) forms the machine learned features and texton based features are considered as hand-crafted features. Random forest (RF) is used to classify the MRI image voxels into normal brain tissues and different parts of tumors, i.e. edema, necrosis and enhancing tumor. The method was evaluated on BRATS 2017 challenge dataset. The results show that the proposed method provides promising segmentations. The mean Dice overlap measure for automatic brain tumor segmentation against ground truth is 0.86, 0.78 and 0.66 for whole tumor, core and enhancing tumor, respectively.