Brain MRI segmentation with patch-based CNN approach

Abstract

Brain Magnetic Resonance Image (MRI) plays a non-substitutive role in clinical diagnosis. The symptom of many diseases corresponds to the structural variants of brain. Automatic structure segmentation in brain MRI is of great importance in modern medical research. Some methods were developed for automatic segmenting of brain MRI but failed to achieve desired accuracy. In this paper, we proposed a new patch-based approach for automatic segmentation of brain MRI using convolutional neural network (CNN). Each brain MRI acquired from a small portion of public dataset is firstly divided into patches. All of these patches are then used for training CNN, which is used for automatic segmentation of brain MRI. Experimental results showed that our approach achieved better segmentation accuracy compared with other deep learning methods.

Ähnliche Arbeiten