Clinical keratoconus (KCN) detection is a challenging and time-consuming task. In the diagnosis process, ophthalmologists must revise demographic and clinical ophthalmic examinations. The latter include slit-lamb, corneal topographic maps, and Pentacam indices (PI). We propose an Ensemble of Deep Transfer Learning (EDTL) based on corneal topographic maps. We consider four pretrained networks, SqueezeNet (SqN), AlexNet (AN), ShuffleNet (SfN), and MobileNet-v2 (MN), and fine-tune them on a dataset of KCN and normal cases, each including four topographic maps. We also consider a PI classifier. Then, our EDTL method combines the output probabilities of each of the five classifiers to obtain a decision based on the fusion of probabilities. Individually, the classifier based on PI achieved 93.1% accuracy, whereas the deep classifiers reached classification accuracies over 90% only in isolated cases. Overall, the average accuracy of the deep networks over the four corneal maps ranged from 86% (SfN) to 89.9% (AN). The classifier ensemble increased the accuracy of the deep classifiers based on corneal maps to values ranging (92.2% to 93.1%) for SqN and (93.1% to 94.8%) for AN. Including in the ensemble-specific combinations of corneal maps’ classifiers and PI increased the accuracy to 98.3%. Moreover, visualization of first learner filters in the networks and Grad-CAMs confirmed that the networks had learned relevant clinical features. This study shows the potential of creating ensembles of deep classifiers fine-tuned with a transfer learning strategy as it resulted in an improved accuracy while showing learnable filters and Grad-CAMs that agree with clinical knowledge. This is a step further towards the potential clinical deployment of an improved computer-assisted diagnosis system for KCN detection to help ophthalmologists to confirm the clinical decision and to perform fast and accurate KCN treatment.
A reduced-order extended state observer (RESO) based a continuous sliding mode control (SMC) is proposed in this paper for the tracking problem of high order Brunovsky systems with the existence of external perturbations and system uncertainties. For this purpose, a composite control is constituted by two consecutive steps. First, the reduced-order ESO (RESO) technique is designed to estimate unknown system states and total disturbance without estimating an available state. Second, the continuous SMC law is designed based on the estimations supplied by the RESO estimator in order to govern the nominal system part. More importantly, the robustness performance is well achieved by compensating not only the lumped disturbance, but also its esti
... Show MoreWireless sensor networks (WSNs) are emerging in various application like military, area monitoring, health monitoring, industry monitoring and many more. The challenges of the successful WSN application are the energy consumption problem. since the small, portable batteries integrated into the sensor chips cannot be re-charged easily from an economical point of view. This work focusses on prolonging the network lifetime of WSNs by reducing and balancing energy consumption during routing process from hop number point of view. In this paper, performance simulation was done between two types of protocols LEACH that uses single hop path and MODLEACH that uses multi hop path by using Intel Care i3 CPU (2.13GHz) laptop with MATLAB (R2014a). Th
... Show MoreThe Electric Discharge (EDM) method is a novel thermoelectric manufacturing technique in which materials are removed by a controlled spark erosion process between two electrodes immersed in a dielectric medium. Because of the difficulties of EDM, determining the optimum cutting parameters to improve cutting performance is extremely tough. As a result, optimizing operating parameters is a critical processing step, particularly for non-traditional machining process like EDM. Adequate selection of processing parameters for the EDM process does not provide ideal conditions, due to the unpredictable processing time required for a given function. Models of Multiple Regression and Genetic Algorithm are considered as effective methods for determ
... Show MoreThe friendly-environment geophysical methods are commonly used in various engineering and near-surface environmental investigations. Electrical Resistivity Imaging technique was used to investigate the subsurface rocks, sediments properties of a proposed industrial site to characterize the lateral and vertical lithological changes. via the electrical resistivity, to give an overview about the karst, weak and robust subsoil zones. Nineteen 2D ERI profiles using Wenner array with 2 m electrode spacing have been applied to investigate the specific industry area. One of these profiles has been conducted with one-meter electrode spacing. The surveyed profiles are divided into a number of blocks, each block consists of several parallel pr
... Show MoreFeature selection (FS) constitutes a series of processes used to decide which relevant features/attributes to include and which irrelevant features to exclude for predictive modeling. It is a crucial task that aids machine learning classifiers in reducing error rates, computation time, overfitting, and improving classification accuracy. It has demonstrated its efficacy in myriads of domains, ranging from its use for text classification (TC), text mining, and image recognition. While there are many traditional FS methods, recent research efforts have been devoted to applying metaheuristic algorithms as FS techniques for the TC task. However, there are few literature reviews concerning TC. Therefore, a comprehensive overview was systematicall
... Show MoreEach organization struggles to exploit each possible opportunity for gaining success and continuing with its work carrier. In this field, organization success can be concluded by fulfilling end user requirements combined with optimizing available resources usage within a specified time and acceptable quality level to gain maximum profit. The project ranking process is governed by the multi-criteria environment, which is more difficult for the governmental organization because other organizations' main target is maximizing profit constrained with available resources. The governmental organization should consider human, social, economic and many more factors. This paper focused on building a multi-criteria optimizing proje
... Show MoreThe most prevalent chronic complication of diabetes mellitus is diabetic neuropathy. The pathogenesis of diabetic neuropathy is exacerbated by hyperglycemia-induced oxidative stress, which causes nerves to deteriorate in a programmed manner. Many clinical trials depend on supplement in an attempt to improve neuropathy symptoms such as (pain & tingling) and patient quality of life, one of them is Coenzyme Q10 which is reported to have an anti-inflammatory and antioxidant effects, and was totally nontoxic and non-reported side effects. This study aimed to evaluate using a Coenzyme Q10 supplement as an adjuvant therapy to gabapentin to improve the clinical symptoms of diabetic neuropathy in relation to its anti-inflammatory and antioxid
... Show MoreThe purpose of this paper is to identify the statistical indicators of the searched variables and identify the relationship between the cognitive learning outcome and the performance of the two mastering skills by parallel spherical standing and equilibrium on the balance beam. And the identification of the percentage of the cognitive learning outcome contribution to the performance of the two mastering skills by parallel spherical standing and the equilibrium on the balance beam. The two researchers used the descriptive approach in the survey method and the correlational relations, being the most appropriate to the nature of the research problem. The research community for the second stage students in the College of Physical Education and
... Show More