LAUSR

Purpose In this work, a novel approach is used to assess macular changes in patients diagnosed with multiple sclerosis. The vast majority of the studies, if not all, do use the capacity of the optical coherence tomography (OCT) to detect changes along the optical path to identify the different retinal layers and the detection of these to measure the thickness of the retina and the different layers. In particular, the measurement of the retinal nerve fibre layer (NFL) has played a fundamental role in related studies. In this work, we computed a fundus image for the NLF, the ganglion cell layer (GCL), the inner plexiform layer, the inner nuclear layer, the outer plexiform layer and the out nuclear layer and analysed the texture of these fundus images to find their relative differences to the healthy control group. Methods A total of 152 eye scans, from 38 healthy controls (age (m/sd): 36.3/9.2 years) and 39 multiple sclerosis patients (age (m/sd): 38.8/7.3 years) were performed by the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA, USA) and exported to be segmented by the OCT Explorer (Retinal Image Analysis Lab, Iowa Institute for Biomedical Imaging, Iowa City, IA, USA). The segmentation was further analysed and corrected when necessary. Results Each of these images was analysed using first and second order statistics. A support vector machine (SVM) classification process was used to assess the capacity to discriminate between healthy controls and patients to find that the layer presenting the highest difference between the control and the patient group is, surprisingly, the GCL, with over 10% in classification accuracy compared to the NFL layer. Conclusions This finding points to the need for further analysis of retinal changes beyond that of the thickness measurement and only in the nerve fiber layer.