Automatic computational scheme for segmentation and identification of acute inflammatory process in multiple sclerosis lesions without using contrast agent

Abstract

Multiple Sclerosis (MS) is an inflammatory demyelinating disease of the Central Nervous System (CNS). It is considered an autoimmune disease in which the immune system wrongly recognizes the myelin sheath of the CNS as an external element and attacks it, resulting in inflammation and scarring (sclerosis) of multiple areas of CNS's white matter substance. The multi-spectral imaging through magnetic resonance (MR) has been successfully used in diagnosing and monitoring MS due to its excellent properties such as high resolution and good differentiation between soft tissues. During a MR exam, it is common to use a gadolinium-based contrast agent to enhance MS lesions undergoing acute inflammation, also known as active lesions. The detection and analysis of such lesions allow the assessment of the treatment prescribed to the patient since it makes possible to analyze the inflammatory stage of the disease. Until recent years we believed that the contrast agent would be completely eliminated from the patient's body after a few days of the administration. However, a number of studies published in 2014 onwards started to indicate that the gadolinium-based contrast agent was being accumulated in certain regions of the brain of patients who had undergone many enhanced MR exams; in this sense, patients with MS fall under this category. The physiological effects of this accumulation are still unknown by the scientific community, which raised serious questions regarding the administration of gadolinium. Given that, since 2015 its usage is being recommended only in cases where it is absolutely necessary. It is also important to note that the use of a gadolinium-based contrast agent also increases the costs and duration of the MR exam. In this scenario, this project proposes the study and development of an automatic computational scheme to perform the segmentation of MS lesions in MR images acquired without contrast agent and indicate, given its texture features, which lesions are in an inflammatory stage. By doing this, MR exams will no longer offer any risk of gadolinium accumulation and its harmful effects. This scheme will also reduce the costs and duration of MR exams. To achieve this goal, finite mixture models, probabilistic anatomical atlases, texture analysis and cortical asymmetry will be investigated and used.