diagnosis of multiple sclerosis from mri images using deep neural network res u-net

Multiple sclerosis (ms) is a chronic inflammatory and degenerative disease of the central nervous system, affecting around 2.8 million people worldwide. the use of advanced techniques of artificial intelligence along with the use of mri images of the brain can be useful for the diagnosis of ms. deep learning algorithms such as convolutional neural networks have outstanding ability to automatically process large amounts of medical images and identify complex relationships in high-dimensional data for disease diagnosis, treatment planning, multiple clinical prediction, and disease control. by examining the performance of the pre-trained deep neural network res u-net, the highest accuracy (99.99 percent) was obtained in the classification of two classes. the findings of the present research can help radiologists to make better and faster decisions in clinical practice due to higher performance. with these automatic models, future studies can show more insights about mri images for ms assessment and thus help in its diagnosis, treatment and control. also, the classification performance of different deep neural network models can be tested by increasing the number of mri images in the data set.