Brain tumor image segmentation based on convolution neural network

Automatic segmentation and early diagnosis of brain tumor is a challenging problem in computer vision and it can provide possibility for pre-operative planning, and solve the problem such as low accurateness and time-consuming in traditional manual segmentation. Under the mentioned problems above, this paper put forward a new method: Based on traditional convolutional neural networks (CNNs), a new architecture model is proposed for automatic brain tumor segmentation, which combines multi-modality images. The newly designed CNNs model automatically learns useful features from multi-modality images to combine multi-modality information. Experiment results show that the proposed model is more accurate than traditional methods and can provide reliable information for clinic treatments.

I. Introduction

After research shows, by this stage in lifestyle and living environment of multiple effects, cancer and related disease incidence is increasing every year. January 25, 2016, the world's top journals in the field of cancer - “CA: Clinicians Journal of Cancer” (CA: A Cancer Journal for Clinicians) published in 2015 cancer statistics released by China Cancer Center Director of China National Cancer Institute Wanqing Chen pointed out that in 2015 China's total cases of cancer incidence were about 4,291,600 and total mortality cases were 2,814,200. According to the American Cancer Society estimates [1], in 2016 the United States will have 1,685,210 new cancer cases, and there will be 595,690 deaths. Especially for adolescents and children (below the age of 19 years), brain tumor has surpassed leukemia and became the biggest killer. Brain tumor is an uncontrolled growth of solid mass formed by undesired cells found in different part of the brain. There are two main types of tumors, benign tumors and malignant tumors. Malignant tumors can be divided into primary tumors and metastatic tumors. Thus, early diagnosis and treatment are important in the Journal for clinical procedure and accurate segmentation of brain tumor becomes the key. Automatic image segmentation technique of brain tumor provides the possibility to solve the problems above.