Preferred Language
Articles
/
fBb2j4oBVTCNdQwCuJ_G
Deep Transfer Learning for Improved Detection of Keratoconus using Corneal Topographic Maps
...Show More Authors
Abstract <p>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.</p>
Scopus Clarivate Crossref
View Publication
Publication Date
Thu Feb 28 2019
Journal Name
Multimedia Tools And Applications
Shot boundary detection based on orthogonal polynomial
...Show More Authors

View Publication
Scopus (40)
Crossref (34)
Scopus Clarivate Crossref
Publication Date
Tue Aug 31 2021
Journal Name
Iraqi Journal Of Science
Plagiarism Detection Methods and Tools: An Overview
...Show More Authors

Plagiarism Detection Systems play an important role in revealing instances of a plagiarism act, especially in the educational sector with scientific documents and papers. The idea of plagiarism is that when any content is copied without permission or citation from the author. To detect such activities, it is necessary to have extensive information about plagiarism forms and classes. Thanks to the developed tools and methods it is possible to reveal many types of plagiarism. The development of the Information and Communication Technologies (ICT) and the availability of the online scientific documents lead to the ease of access to these documents. With the availability of many software text editors, plagiarism detections becomes a critical

... Show More
View Publication Preview PDF
Scopus (29)
Crossref (17)
Scopus Crossref
Publication Date
Fri Jul 21 2023
Journal Name
Journal Of Engineering
ESTIMATION OF THE OPTIMUM BED THICKNESS OF A FLOWTHROUGH POROUS ELECTRODE (FTPE) WORKING UNDER MASS TRANSFER CONTROL
...Show More Authors

In this paper, a theoretical analysis of optimum bed thickness operates under mass transfer control for realizing a high efficiency and reaction conversion of an electrochemical reactor has been made based on flowthrough porous electrode (FTPE) configuration. Many models have been used to represent the optimum bed thickness by taking a look into previous works concerned and collecting all related information, data, and models. The parameters that affect the optimum bed thickness have been visualized and reviewed, and almost all of them have been examined by experimental data from different sources and based on the various models. It has been found that the increase in electrolyte flow rate, concentration, limiting current density, and sp

... Show More
View Publication Preview PDF
Crossref (2)
Crossref
Publication Date
Fri Jan 26 2024
Journal Name
Iraqi Journal Of Science
Physical–Spectroscopic Study of Charge-Transfer Complexes of Some Purine Derivatives with (π) and (σ) Electron Acceptors
...Show More Authors

Charge-transfer (CT) complexes of adenine (Ade.), guanine (Gua.), xanthine (Xan.), and inosine (Ino.) as electron donors with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), 2,3,5,6-tetrabromo-1,4-benzoquinone (Bromanil)(BA) as π – electron acceptors and iodine (Iod.) as σ – electron acceptor were studied and their electronic spectra recorded .In each case one (CT) band was observed and recorded. These spectroscopic investigations made in ethanol solvent at (20°C) temperature. The values of equilibrium constant (KCT), change in standard free energy (ΔG°), molar extinction coefficient (εCT(, absorption band energy (hνCT) of CT complexes and the association energy of the CT complexes-excited state (W) were calculated and studie

... Show More
View Publication Preview PDF
Publication Date
Sun Apr 08 2018
Journal Name
Al-khwarizmi Engineering Journal
Experimental Investigation of the Effect of Curvature Ratio on Heat Transfer in Double Pipe Helical Heat Exchanger
...Show More Authors

Different parameters of double pipe helical coil were investigation experimentally. Four coils were used; three with a curvature ratio (0.037, 0.031, and 0.028) and 11mm diameter of the inner tube while the fourth with 0.033 curvature ratio and 13 mm diameter of the inner tube. The hot water flow in the inner tube whereas the cold water flows in the annulus. The inlet temperatures of hot and cold water are 50 0C and 18 0C respectively. The inner mass flow rate ranges from 0.0167 to 0.0583 kg/s. The results show the Nusselt number increase with increase curvature ratio. The Nusselt number of the coil with 0.037 curvature ratio increases by approximately 12.3 % as compare with 0.028 curvature ratio. The results also r

... Show More
View Publication Preview PDF
Crossref
Publication Date
Sat Jul 01 2023
Journal Name
Journal Of Accounting And Financial Studies ( Jafs )
Analyzing impact of competitive dimensions on the efficiency of e-learning: دراسه استطلاعيه
...Show More Authors

The aim of this research is to diagnose the impact of competitive dimensions represented by quality, cost, time, flexibility on the efficiency of e-learning, The research adopted the descriptive analytical method by identifying the impact of these dimensions on the efficiency of e-learning, as well as the use of the statistical method for the purpose of eliciting results. The research concluded that there is an impact of the competitive dimensions on the efficiency of e-learning, as it has been proven that the special models for each of the research hypotheses are statistically significant and at a level of significance of 5%, and that each of these dimensions has a positive impact on the dependent variable, and the research recommended

... Show More
View Publication Preview PDF
Publication Date
Fri Dec 30 2011
Journal Name
Al-kindy College Medical Journal
The Role of the Use of Low Molecular Weight Heparin in the Prevention of Deep Venous Thrombosis after Total Knee Arthroplasty
...Show More Authors

Background A prospective clinical study was
performed to compare the efficacy of the use of lowmolecular-
weight heparin group (enoxparin group)
with control group in the prevention of deep-vein
thrombosis after total knee arthroplasty.
Aim of the study: to assess the prevalence of DVT
after total knee arthroplasty and evaluate the
importance of the use of low molecular weight
heparin in the prevention of this DVT.
Methods Thirty-three patients undergoing total
knee arthroplasty were randomly divided into two
groups. One group consisted of 12 patients who
received no prophylaxis with an anticoagulant (the
control group), other group consisted of 21 patients
who received the low-molecular-weight h

... Show More
View Publication Preview PDF
Publication Date
Mon Apr 04 2022
Journal Name
Journal Of Educational And Psychological Researches
Motivated Classroom Evaluation Environment for Learning and its Relationship to Feedback Patterns among Teachers of the Second Cycle of Basic Education in the Sultanate of Oman
...Show More Authors

This study aims to investigate the degree of practicing the motivated classroom evaluation environment for learning and its relationship to different feedback patterns. To achieve the objectives of the study, the correlational descriptive research design was employed. A questionnaire was constructed consisting of two parts: the classroom evaluation environment (13) items, and feedback patterns (24) items on a five-point scale. The psychometric properties of the questionnaire were verified in terms of validity and reliability. The questionnaire was applied to a sample of (265) male and female teachers who work in the second cycle schools for grades (5-10) of basic education in all academic majors in the Governorate of Muscat in the Sultan

... Show More
View Publication Preview PDF
Publication Date
Sun Sep 03 2023
Journal Name
Iraqi Journal Of Computers, Communications, Control & Systems Engineering (ijccce)
Efficient Iris Image Recognition System Based on Machine Learning Approach
...Show More Authors

HM Al-Dabbas, RA Azeez, AE Ali, IRAQI JOURNAL OF COMPUTERS, COMMUNICATIONS, CONTROL AND SYSTEMS ENGINEERING, 2023

View Publication
Publication Date
Sat Apr 15 2023
Journal Name
Iraqi Journal Of Science
Best Way to Detect Breast Cancer by UsingMachine Learning Algorithms
...Show More Authors

Breast cancer is the second deadliest disease infected women worldwide. For this
reason the early detection is one of the most essential stop to overcomeit dependingon
automatic devices like artificial intelligent. Medical applications of machine learning
algorithmsare mostly based on their ability to handle classification problems,
including classifications of illnesses or to estimate prognosis. Before machine
learningis applied for diagnosis, it must be trained first. The research methodology
which isdetermines differentofmachine learning algorithms,such as Random tree,
ID3, CART, SMO, C4.5 and Naive Bayesto finds the best training algorithm result.
The contribution of this research is test the data set with mis

... Show More
View Publication Preview PDF