No abstract

Keratoconus (KC) is a type of eye disease that involves the thinning of the corneal layer and a change in the semispherical shape of the cornea to a bulging cone shape when viewed laterally. KC is difficult to detect in the early stages of the disease, as the patient does not feel any pain. Hence, the development of a KC detection (KD) method using a digital image processing approach is needed for the early detection of KC so that physicians can provide patients with the subsequent treatment sooner. The objective of this study was to develop a method of KD using a camera from a smart device to capture anterior and lateral segment photographed eye images (A&LSPIs). A total of 280 images comprising 140 KC and 140 normal A&LSPIs were used in this study, and all images were validated by a qualified expert. First, the corneal area of both image views was segmented so the geometric features could be extracted using the automated modified active contour model and the semiautomated spline function for the anterior and lateral images, respectively. Then, the features were selected using infinite latent feature selection (ILFS) by identifying the feature rankings based on the graph weighting that was automatically learned by the probabilistic latent semantic analysis (PLSA). The results showed that the all-combined features, where the proposed and improved features were successfully top ranked, had 96.05% accuracy, 98.41% sensitivity and 93.65% specificity with the RFn=50 classifier, outperforming the 7-NNMaha and SVMRBF classifiers. In conclusion, this study successfully developed a keratoconus detection system based on fusion features using a digital image processing approach for A&LSPIs captured with a smartphone camera.INDEX TERMS Keratoconus, lateral segment photographed images, anterior segment photographed images, fusion features, digital image processing

No abstract

Keratoconus (KC) is a condition of the bulging of the eye cornea. It is a common non-inflammatory ocular disorder that affects mostly the younger populace below the age of 30. The eye cornea bulges because of the conical displacement of either outwards or downwards. Such condition can greatly reduce one’s visual ability. Therefore, in this paper, we afford a mobile solution to mitigate the KC disorder using the state-of-the-art deep transfer learning method. We intend to use the pre-trained VGGNet-16 model and a conventional convolutional neural network to detect KC automatically. The experimental work uses a total of 4000 side view lateral segment photographed images (LSPIs) comprising 2000 of KC and non-KC or healthy each involving 125 subjects. The LSPIs were extracted from the video data captured using a smartphone. Fine tuning of three hyperparameters namely the learning rate (LR), batch size (BS) and epoch number (EN) were carried out during the training phase to generate the best model of which, the VGGNet-16 model fulfilled it. For the KC detection task, our proposed model achieves an accuracy of 95.75%, a sensitivity of 92.25%, and specificity of 99.25% using the LR, BS and EN of 0.0001, 16, and 70, respectively. These results confirmed the high potential of our proposed solution to apprehend KC prevalence towards an automated KC screening procedure.

No abstract