Abstract
Magnetic resonance imaging (MRI) is a clinically important tool for diagnosing several neurological diseases such as the multiple sclerosis (MS). Brain MRI has always facilitated the examination of the MS pathology. Moreover, spinal cord MRI is greatly suggested for the management of such disease, even though the use of conventional spinal cord MRI can be a challenging task. In fact, it is a long and fine organ that has some mobility and that suffers from breathing artifacts, low contrast, heartbeat and cerebro-spinal fluid flows. In this study, to identify spinal cord damage in MS patient, an adaptive MRI contrast enhancement (CE), called the LL-GAGC method, is proposed. This novel technique is based on a combination of the adaptive gamma correction and the discrete wavelet transform with singular value decomposition algorithms. The main reason of this association is to enhance adaptively the contrast of dark MR images while preserving edge information from any distortion. A large database formed by 112 T2-w spinal cord MR images was examined for assessment of the proposed LL-GAGC CE method. Qualitative and quantitative evaluations demonstrated that our proposed algorithm performs well in enhancing the contrast of dark MR images with the benefit of preserving brightness information and edges details.