Data Preprocessing via Multi-sequences MRI Mixture to Improve Brain Tumor Segmentation

Automatic brain tumor segmentation is one of the crucial problems nowadays among other directions and domains where daily clinical workflow requires to put a lot of efforts while studying computer tomography (CT) or structural magnetic resonance imaging (MRI) scans of patients with various pathologies. The MRI is the most common method of primary detection, non-invasive diagnostics and a source of recommendations for further treatment. The brain is a complex structure, different areas of which have different functional significance. In this paper, we propose a robust pre-processing technique which allows to consider all available information from MRI scans by composition of T1, T1C and FLAIR sequences in the unique input. Such approach enriches the input data for the automatic segmentation process and helps to improve the accuracy of the segmentation performance. Proposed method demonstrates significant improvement on the binary segmentation problem with respect to Dice and Recall metrics compare to similar training/evaluation procedure based on any single sequence regardless of the chosen neural network architecture. Obtained results demonstrates significant evaluation improvement while combining three MRI sequences either as weighted mixture to get 1-channel mixed up image or in the 3-channel RGB like image for both considered problems - binary brain tumor segmentation with and without inclusion of edema in the region of interest (ROI). Final improvements on the test part of data set are in the range of 5.6-9.1% on the single-fold trained model according to the Dice metric with the best value of 0.902 without considering a priori "empty" slides. We also demonstrate strong impact on the Recall metric with the growth up to 9.5%. Additionally this approach demonstrates significant improvement according to the Recall metric getting the increase by up to 11%.