Deep Neural Network-Based Ensemble Model for Eye Diseases Detection and Classification

Abstract

Fundus images are the principal tool for observing and recognizing a wide range of ophthalmological abnormalities. The automatic and robust methods based on color fundus images are urgently needed since few symptoms are observable in the early stages of the disease. Experts must manually evaluate images to detect diseases for screening procedures to be effective. Due to the complexity of the screening procedure and the shortage of experienced personnel, developing successful screening-based treatments is costly. Although existing automated approaches strive to address these issues, they cannot handle a wide range of diseases and real-world circumstances. We design an automated deep learning-based ensemble method to detect and classify eye diseases from fundus images to address the abovementioned problems. A deep CNN-based model is proposed in the ensemble method that incorporates a mix of 20 layers, including the activation, optimization, and loss functions. The contrast-limited adaptive histogram equalization (CLAHE) and Gaussian filter are utilized in the pre-processing step to get more explicit images and eliminate noise. To avoid overfitting in the training phase, augmentation techniques are applied. Three pre-trained CNN models, including VGG16, DenseNet201, and ResNet50, are employed to compare and assess the efficiency of the proposed CNN model. Experimental results demonstrate that the ensemble approach outperforms recent approaches, which is comparatively state-of-art in the ODIR publicly available dataset.

SIMD implementation of deep CNNs for myopia detection on a single-board computer system

“…This implementation platform is characterized by its deficient power consumption compared to other implementation platforms shown in Table 2, where a set of proposed systems that were presented as aids in detecting and diagnosing ocular diseases are highlighted in this table, wherein the results in Fig. 9-11 show that the power gain achieved in the proposed system compared to the systems presented in the same field alternated between 72.6 %, 68.49 %, 19.17 %, and 47.94 % for [34][35][36][37] respectively in Table 2.…”

“…The comparisons reviewed in Table 2 show a clear superiority of the presented deep network and the proposed system over its counterparts of neural networks in previous and presented studies for ocular disease detection and diagnosing. The table also indicates the high accuracy concluded by the proposed network relative to its limited size, which may reach 18.667 and 3, compared to those in [36,37]. The size of our network stands out with the possibility of implementing it on various implementation platforms with limited power consumption as a single-board mobile system, with a maximum power consumption of about 3.65 W, compared to those vast networks that can only be implemented on GPU with high specifications and significant power consumption, which is part of an integrated computer system with a very high cost.…”