A lesion is an area of tissue that has been damaged through injury or disease. So a brain lesion is an area of injury or disease within the brain. While the definition sounds simple, understanding brain lesions can be complicated. That's because there are many types of brain lesions. They can range from small to large, from few to many, from relatively harmless to life-threatening. magnetic resonance imaging (MRI) is increasingly used nowadays. Manual delineation of Multiple Sclerosis (MS) lesions in MR images by human experts is time-consuming, subjective and prone to inter-expert variability. Therefore, automatic segmentation is needed as an alternative to manual segmentation. In this paper, the 2D discrete wavelet transform (DWT) is used to extract local information from analyzing MR images. Ensemble decision trees (EDT) and Ensemble support vector machines (ESVM) are used to segment MS lesions and automatically differentiate between blocks in regions of MS lesions and the blocks in non-MS. lesions. We evaluate our approach on real MRI data sets. We can detect MS lesions with accuracy more than 98 %. Technique evaluated using real MRI datasets. The results compared with ground truth. The main contribution of the proposed techniques described in this paper is the use of DWT with Ensemble Machine Learning and solving the problem of imbalanced classification data without changing or losing trained data.