Computer-aided diagnosis of keratoconus through VAE-augmented images using deep learning.

Detecting clinical keratoconus (KCN) poses a challenging and time-consuming task. During the diagnostic process, ophthalmologists are required to review demographic and clinical ophthalmic examinations in order to make an accurate diagnosis. This study aims to develop and evaluate the accuracy of deep convolutional neural network (CNN) models for the detection of keratoconus (KCN) using corneal topographic maps. We retrospectively collected 1758 corneal images (978 normal and 780 keratoconus) from 1010 subjects of the KCN group with clinically evident keratoconus and the normal group with regular astigmatism. To expand the dataset, we developed a model using Variational Auto Encoder (VAE) to generate and augment images, resulting in a dataset of 4000 samples. Four deep learning models were used to extract and identify deep corneal features of original and synthesized images. We demonstrated that the utilization of synthesized images during training process increased classification performance. The overall average accuracy of the deep learning models ranged from 99% for VGG16 to 95% for EfficientNet-B0. All CNN models exhibited sensitivity and specificity above 0.94, with the VGG16 model achieving an AUC of 0.99. The customized CNN model achieved satisfactory results with an accuracy and AUC of 0.97 at a much faster processing speed compared to other models. In conclusion, the DL models showed high accuracy in screening for keratoconus based on corneal topography images. This is a development toward the potential clinical implementation of a more enhanced computer-aided diagnosis (CAD) system for KCN detection, which would aid ophthalmologists in validating the clinical decision and carrying out prompt and precise KCN treatment.

Human endogenous retrovirus protein activates innate immunity and promotes experimental allergic encephalomyelitis in mice.

Multiple sclerosis (MS) is a complex multifactorial disease of the central nervous system (CNS) for which animal models have mainly addressed downstream immunopathology but not potential inducers of autoimmunity. In the absence of a pathogen known to cause neuroinflammation in MS, Mycobacterial lysate is commonly used in the form of complete Freund's adjuvant to induce autoimmunity to myelin proteins in Experimental Allergic Encephalomyelitis (EAE), an animal model for MS. The present study demonstrates that a protein from the human endogenous retrovirus HERV-W family (MSRV-Env) can be used instead of mycobacterial lysate to induce autoimmunity and EAE in mice injected with MOG, with typical anti-myelin response and CNS lesions normally seen in this model. MSRV-Env was shown to induce proinflammatory response in human macrophage cells through TLR4 activation pathway. The present results demonstrate a similar activation of murine dendritic cells and show the ability of MSRV-Env to trigger EAE in mice. In previous studies, MSRV-Env protein was reproducibly detected in MS brain lesions within microglia and perivascular macrophages. The present results are therefore likely to provide a model for MS, in which the upstream adjuvant triggering neuroinflammation is the one detected in MS active lesions. This model now allows pre-clinical studies with therapeutic agents targeting this endogenous retroviral protein in MS.