It can be summarized that the main objective is to develop a method to assist brain tumor segmentation which works in the same line of work of a physician
It can be summarized that the main objective is to develop a method to assist brain tumor segmentation which works in the same line of work of a physician, considering his experience and knowledge. Various techniques were proposed for segmenting an MRI image which comparatively take lesser time than manual operations to detect and extract the brain tumor. Deep learning architecture is rising computational paradigm in developing predictive models of the diseases. In our study, after conducting literature review of various techniques related to the classification and analysis of Biomedical images for detection of Brain tumor, it has been observed that deep neural networks have been very much accurate and automatic in diagnoses of diseases of patients. So have been helpful for providing proper treatment to save precious human life.
There is no universally accepted method for image segmentation, as of the result of image segmentation is affected by lots of factors. Thus there is no single method which can be considered good as all the methods are equally good for a particular type of image. Due to this, image segmentation remains a challenging problem in image processing.
The future work is to improve the classification accuracy by extracting more features and increasing the training data set and also the method can be produced that provide very good results in enhancing, detecting and segmenting the brain tumor from a MR images. Besides the energy, correlation, contrast and homogeneity add more information to the feature extraction in order to make the system more sensitive information from the textures or location. It will be interesting to continue developing more adaptive models for other types of brain tumors. It should be clear that many factors influence the appearance of tumors on images, and although there are some common features of malignancies, there is also a great deal of variation that depends on the tissue and the tumor type. Further in future, we will focus on constructing even deeper convolutional neural networks for biomedical applications with even more improved training time and accuracy.