Convolution neural network based brain tumour detection using efficient classification technique

Image processing has become an essential component in many fields of biomedical research such as tumor detection, automatically determining the volume of a heart chamber, screening brain scans for possible diseases. Different techniques for automatic detection of brain tumor involve various steps: image acquisition, segmentation, classification using neural network and optimization, and identification of tumor type. This paper presents a new approach to detect and segment brain tumors. The detection and segmentation of brain tumors can be formulized as novelty detection by using Hybrid probability based straightened bound segmentation model. The main objective of the proposed method is to precisely identify the presence of tumour cells in brain images as an early indication of malignant cells that may cause to the demise of patients.