Neethu Ouseph, Mrs Shruti
2017 unpublished
In current days a proficient detection of brain tumor is being breathtaking challenge in medical field. An Automatic segmentation of brain images has a significant role in reducing the difficulties of manual labeling and increasing the strength of brain tumor diagnosis. The manual segmentation can lead to intra and inter errors. Image segmentation techniques are help to get the meaningful information that are useful in the detection of tumor. Magnetic resonance imaging (MRI) has a high spatial
more » ... has a high spatial resolution view of brain and it is a very powerful tool used to diagnose a wide range of disorders and it has been proven to be a highly flexible imaging technique. This paper presents a robust detection and extraction method based on Artificial Neural Network that reduces operators and errors. Artificial Neural Networks (ANNs) are mathematical analogues of biological neural systems. This system is made up of a parallel interconnected system of nodes; called neurons. The image processing techniques such as image conversion, feature extraction, bias field correction and histogram equalization have been developed for extraction of the tumor the MRI images of the cancer affected patients. A Fuzzy c means Classifier is developed to recognize healthier tissue from cancer tissue. The project is divided into two phases: Training Phase and Testing Phase. The aim of the project is to detect and extract the tissue that contains abnormalities. The specificity and the sensitivity of the method is evaluated and accuracy is determined. The performance parameters show significant outputs which are helpful in extracting tumor from brain MRI image.