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 steganography hides information within other information, such as file, message, picture, or video. A cryptography is the science of converting the information from a readable form to an unreadable form for unauthorized person. The main problem in the stenographic system is embedding in cover-data without providing information that would facilitate its removal. In this research, a method for embedding data into images is suggested which employs least significant bit Steganography (LSB) and ciphering (RSA algorithm) to protect the data. System security will be enhanced by this collaboration between steganography and cryptography.
The Purpose of this Research show gap between a Normal Cost System and Resource consumption Accounting Applied in AL-Rafidin Bank.
The Research explores that, how the idle capacity can be determined under resource consumption accounting, discuss the possibility of employing these energies. Research also viewed how costs can be separated into Committee and Attribute. Resource Consumption Accounting assists managers in pricing services or products based on what these services or products use from each Source.
This Research has been proven
This paper seeks to study the link between the fundamentalist evidence based on the observance of governance and interests and the ranks of the three legitimate purposes (necessary, need and detailed). The researcher followed the descriptive-analytical approach. The study reached important results, including that the measurement relates to the three ranks, but predominantly attached to measure the meaning of the need and the need, and the measurement of the semi-formal and semi-predominance improvement. Reclamation is considered by the majority of scholars to be authentic if it is related to the necessity and the need, and that it is not acceptable to improve only by a witness who recommends it. The excuses relate to Hajji and Tahini, no
... Show MoreThis paper includes an experimental study of hydrogen mass flow rate and inlet hydrogen pressure effect on the fuel cell performance. Depending on the experimental results, a model of fuel cell based on artificial neural networks is proposed. A back propagation learning rule with the log-sigmoid activation function is adopted to construct neural networks model. Experimental data resulting from 36 fuel cell tests are used as a learning data. The hydrogen mass flow rate, applied load and inlet hydrogen pressure are inputs to fuel cell model, while the current and voltage are outputs. Proposed model could successfully predict the fuel cell performance in good agreement with actual data. This work is extended to developed fuel cell feedback
... Show Morethe research ptesents a proposed method to compare or determine the linear equivalence of the key-stream from linear or nonlinear key-stream
Recently The problem of desertification and vegetation cover degradation become an environmental global challenge. This problem could be summarized as as the land cover changes. In this paper, the area of Al- Muthana in the south of Iraq will be consider as one of Semi-arid lands. For this purpose, the Ladsat-8 images can be used with 15 m in spatial resolution. In order to over Achieve the work, many important ground truth data must be collected such as, rain precipitation, temperature distribution over the seasons, the DEM of the region, and the soil texture characteristics. The extracted data from this project are tables, 2-D figures, and GIS maps represent the distributions of vegetation areas, evaporation / precipitation, river levels
... Show MoreIntroduction: All-ceramic crowns are widely used in prosthodontics and cosmetic dentistry due to their good esthetic and proper physical properties. Chipping of ceramic is one of the most common post-insertion complications, that can be fixed either extraoral or intraorally. The latter is time time-effective alternative, less traumatic, and low-cost. A newer objective method of laser is a surface modification of ceramics to increase surface roughness. The aim of this study is to provide a review of Er,Cr;YSGG (2960nm) in intraoral repair and shear bond strength (SBS). Method: A thorough search considering Google Scholar and PubMed published data and ten articles found wh
... Show MoreThe nonlinear refractive index and the nonlinear absorption coefficient of unmodified and functional poly(methyl methacrylate) PMMA films were studied before and after the addition of the filler by the z-scan technique, using a Q-switched Nd:YAG laser at two wavelengths: 532 nm and 1064 nm, and at three input energies (13, 33 and 53) mJ. Both linear and nonlinear refractive indices and absorption coefficients of polymer films were studied by using UV-VIS spectrophotometer. The results show that the creation of functional PMMA from unmodified PMMA will increase the nonlinear optical properties in the functional PMMA/copper matrix more than in the unmodified PMMA/copper matrix. Hence, the functional PMMA appears promising as a useful third
... Show MoreThe main purpose of the paper is to identify the controllability of an existing production system; yogurt production line in Abu Ghraib Dairy Factory which has several machines of food processing and packing that has been studied. Through the starting of analysis, instability in production has been found in the factory. The analysis is built depending on experimental observation and data collection for different processing time of the machines, and statistical analysis has been conducted to model the production system. Arena Software is applied for simulating and analyzing the current state of the production system, and results are expanded to improve the system production and efficiency. Research method is applied to contribute in knowi
... Show MoreThis research estimates the effect of independent factors like filler (3%, 6%, 9%, 11% weight fraction), normal load (5N, 10N, 15N), and time sliding (5,7 , 9 minutes) on wear behavior of unsaturated polyester resin reinforced with jute fiber and waste eggshell and, rice husk powder composites by utilizing a statistical approach. The specimens polymeric composite prepared from resin unsaturated polyester filled with (4% weight fraction) jute fiber, and (3%, 6%, 9%, 11% weight fraction) eggshell, and rice husk by utilizing (hand lay-up) molding. Dry sliding wear experiments were carried utilizing a standard (pin on disc test setup) following a well designed empirical schedule that depends on Taguchi’s experimental design L9 (MINIT
... Show More