Deep Learning for Multiple Sclerosis (DeepMS)

Project no.: PP-88J/19

Project description:

Multiple sclerosis is an autoimmune disease of the central nervous system. It damages myelin sheaths of neurons and manifest itself in relapsing-remitting or progressing forms. In the long term, it may cause certain disabilities. Magnetic resonance tomography imaging is the only visual test used for the diagnosis of multiple sclerosis. It is a non-invasive test without the use of damaging radiation. This test enables the analysis of characteristic white matter lesions caused by multiple sclerosis. Also, it allows to assess changes over time, the number, size and location of lesions, as well as activity of gadolinium-enhanced lesions centres. The assessment of the mentioned aspects is a very complicated process, which takes a long time, is expensive and subjective. In order to achieve a faster and more objective assessment of lesions in multiple sclerosis patients, an effective and accurate recognition and segmentation system are necessary. During this project, a representative sample of MRI scans of MS patients will be obtained, and it will be used construct an artificial intelligence-based system for the assessment fo lesions multiple sclerosis patients. Such a system will enable faster and more accurate diagnosis of multiple sclerosis and assessment of its stage and development over time. To achieve these goals, deep learning methods will be used, and additionally, capsule neural network will be developed in order to gain more accuracy.

Project funding:

KTU R&D&I Fund

Period of project implementation: 2019-04-01 - 2019-12-31

Project partners: Lithuanian University of Health Sciences

Tomas Iešmantas

2019 - 2019

Department of Applied Mathematics, Faculty of Mathematics and Natural Sciences

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more or switch them off if you prefer. However, by continuing to use the site without changing settings, you are agreeing to our use of cookies.
I agree