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.
The current research aims to identify the effect of the learning mastery strategy using interactive learning as a therapeutic method on the achievement of secondary school students in mathematics. To achieve the research objective, the researcher selected second-grade middle school students at Al-Haybah Intermediate School for Boys and determined his research sample, which consisted of (77) students distributed into two sections: Section (A) the experimental group, with (38) students, and Section (B) the control group, with (39) students. The statistical equivalence of the two research sample groups was confirmed in the variables (intelligence test, previous achievement, and previous knowledge test). The researchers chose the par
... Show MoreA simple, rapid, sensitive and inexpensive approach is described in this work based on a combination of solid‐phase extraction of 8‐hydroxyquinoline (8HQ), for speciation and preconcentration of Cr(III) and Cr(VI) in river water, and the direct determination of these species using a flow injection system with chemiluminescence detection (FI–CL) and a 4‐diethylamino phenyl hydrazine (DEAPH)–hydrogen peroxide system. At different pH, the two forms of chromium [Cr(III) and Cr(VI)] have different exchange capacities for 8HQ, therefore two columns were constructed; the pH of column 1 was adjusted to pH 3 for retaining Cr(III) and column 2 was adjusted to pH 1 for retaining of Cr(VI). The sorbe
The aims of this study are to explore the commercial artifacts in the following three kinds of vegetables oils, Nigella Sativa, Trigonella foenum-graecum Linn,and Zingiber officinale. These oils have been very popular medicinal plants which are commonly used in traditional medicine .These commercial oils have been compared with the extracts of these plants.
The physical properties of extracts and commercial oils of these plants have been stuied. We observed that the refractive index of the plants matches and non-significant, while specific gravity of Nigella Sativa has similar specific gravity in both extracts and commercial oil in contrast with Trigonella foenum Linn,and Zingiber officinale and we found significant difference (P<
Imitation learning is an effective method for training an autonomous agent to accomplish a task by imitating expert behaviors in their demonstrations. However, traditional imitation learning methods require a large number of expert demonstrations in order to learn a complex behavior. Such a disadvantage has limited the potential of imitation learning in complex tasks where the expert demonstrations are not sufficient. In order to address the problem, we propose a Generative Adversarial Network-based model which is designed to learn optimal policies using only a single demonstration. The proposed model is evaluated on two simulated tasks in comparison with other methods. The results show that our proposed model is capable of completing co
... Show MoreECG is an important tool for the primary diagnosis of heart diseases, which shows the electrophysiology of the heart. In our method, a single maternal abdominal ECG signal is taken as an input signal and the maternal P-QRS-T complexes of original signal is averaged and repeated and taken as a reference signal. LMS and RLS adaptive filters algorithms are applied. The results showed that the fetal ECGs have been successfully detected. The accuracy of Daisy database was up to 84% of LMS and 88% of RLS while PhysioNet was up to 98% and 96% for LMS and RLS respectively.