Performance evaluation of brain tumor detection using watershed Segmentation and thresholding

Abstract Brain tumors and cancers are life-threatening diseases to human beings and have been on the rise. If undetected, they are deadly. With the advent of advanced medical technology, it has become imperative to accurately spot and identify these tumors at the earliest. The manuscript aims at providing an accurate method to detect and segment brain tumors from MRI scans. This is achieved by implementing watershed segmentation and threshold algorithm paired with pre and post image processing techniques. Apart from detecting the tumor region, the proposed process also enhances image quality by noise removal techniques and image quality improvement. These results give promising values when verified using several evaluation parameters such as Structural Similarity Index Measure (SSIM), Feature Similarity Index Measure (FSIM) and Peak Signal-to-Noise Ratio (PSNR) and stand out among the other similar pre-existing algorithms that they are compared with in a comparative analysis.