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
Sun Oct 01 2017
Journal Name
Ieee Transactions On Neural Systems And Rehabilitation Engineering
A Framework of Temporal-Spatial Descriptors-Based Feature Extraction for Improved Myoelectric Pattern Recognition
...Show More Authors

View Publication
Scopus (123)
Crossref (124)
Scopus Clarivate Crossref
Publication Date
Tue Jan 17 2017
Journal Name
International Journal Of Science And Research (ijsr)
Detection System of Varicose Disease using Probabilistic Neural Network
...Show More Authors

Publication Date
Wed Aug 05 2020
Journal Name
Advances In Structural Engineering
Strength compensation of deep beams with large web openings using carbon fiber–reinforced polymer sheets
...Show More Authors

This article presents the results of an experimental investigation of using carbon fiber–reinforced polymer sheets to enhance the behavior of reinforced concrete deep beams with large web openings in shear spans. A set of 18 specimens were fabricated and tested up to a failure to evaluate the structural performance in terms of cracking, deformation, and load-carrying capacity. All tested specimens were with 1500-mm length, 500-mm cross-sectional deep, and 150-mm wide. Parameters that studied were opening size, opening location, and the strengthening factor. Two deep beams were implemented as control specimens without opening and without strengthening. Eight deep beams were fabricated with openings but without strengthening, while

... Show More
View Publication
Crossref (19)
Crossref
Publication Date
Wed Aug 05 2020
Journal Name
Advances In Civil Engineering
Strength compensation of deep beams with large web openings using carbon fiber–reinforced polymer sheets
...Show More Authors

This article presents the results of an experimental investigation of using carbon fiber–reinforced polymer sheets to enhance the behavior of reinforced concrete deep beams with large web openings in shear spans. A set of 18 specimens were fabricated and tested up to a failure to evaluate the structural performance in terms of cracking, deformation, and load-carrying capacity. All tested specimens were with 1500-mm length, 500-mm cross-sectional deep, and 150-mm wide. Parameters that studied were opening size, opening location, and the strengthening factor. Two deep beams were implemented as control specimens without opening and without strengthening. Eight deep beams were fabricated with openings but without strengthening, while

... Show More
Scopus (19)
Crossref (19)
Scopus Clarivate Crossref
Publication Date
Mon Oct 31 2022
Journal Name
International Journal Of Intelligent Engineering And Systems
Robot Path Planning in Unknown Environments with Multi-Objectives Using an Improved COOT Optimization Algorithm
...Show More Authors

Scopus (9)
Crossref (2)
Scopus Crossref
Publication Date
Sun Jan 01 2023
Journal Name
Journal Of Interdisciplinary Mathematics
On maps of Γ -period 2 on prime Γ- near-rings
...Show More Authors

We present the concept of maps Γ- periodi2 on Γ -near-ring S. Our main goal is to research and explore the presence and mapping traits such as h Γ –hom anti-Γ –hom, Γ –α-derivations of Γ -periodi2 on Γ- near-rings.

View Publication
Scopus Clarivate Crossref
Publication Date
Wed Jul 01 2020
Journal Name
Journal Of Physics: Conference Series
Improved Alternating Direction Implicit Method
...Show More Authors
Abstract<p>The alternating direction implicit method (ADI) is a common classical numerical method that was first introduced to solve the heat equation in two or more spatial dimensions and can also be used to solve parabolic and elliptic partial differential equations as well. In this paper, We introduce an improvement to the alternating direction implicit (ADI) method to get an equivalent scheme to Crank-Nicolson differences scheme in two dimensions with the main feature of ADI method. The new scheme can be solved by similar ADI algorithm with some modifications. A numerical example was provided to support the theoretical results in the research.</p>
View Publication
Scopus (1)
Crossref (1)
Scopus Crossref
Publication Date
Wed Feb 15 2023
Journal Name
Full Text Book Of Minar Congress 7
EVALUATING THE CHANGE DETECTION OF(NDVI) FOR BABYLON CITY USING REMOTE SENSING AND GIS TECHNIQUES (2015-2020)
...Show More Authors

The normalized difference vegetation index (NDVI) is an effective graphical indicator that can be used to analyze remote sensing measurements using a space platform, in order to investigate the trend of the live green vegetation in the observed target. In this research, the change detection of vegetation in Babylon city was done by tracing the NDVI factor for temporal Landsat satellite images. These images were used and utilized in two different terms: in March 19th in 2015 and March 5th in 2020. The Arc-GIS program ver. 10.7 was adopted to analyze the collected data. The final results indicate a spatial variation in the (NDVI), where it increases from (1666.91 𝑘𝑚2) in 2015 to (1697.01 𝑘𝑚2)) in 2020 between the t

... Show More
View Publication
Crossref
Publication Date
Wed Jan 01 2020
Journal Name
Technologies And Materials For Renewable Energy, Environment And Sustainability: Tmrees20
Change detection of the land cover for three decades using remote sensing data and geographic information system
...Show More Authors

View Publication
Crossref (3)
Crossref
Publication Date
Sat Apr 30 2022
Journal Name
Revue D'intelligence Artificielle
Performance Evaluation of SDN DDoS Attack Detection and Mitigation Based Random Forest and K-Nearest Neighbors Machine Learning Algorithms
...Show More Authors

Software-defined networks (SDN) have a centralized control architecture that makes them a tempting target for cyber attackers. One of the major threats is distributed denial of service (DDoS) attacks. It aims to exhaust network resources to make its services unavailable to legitimate users. DDoS attack detection based on machine learning algorithms is considered one of the most used techniques in SDN security. In this paper, four machine learning techniques (Random Forest, K-nearest neighbors, Naive Bayes, and Logistic Regression) have been tested to detect DDoS attacks. Also, a mitigation technique has been used to eliminate the attack effect on SDN. RF and KNN were selected because of their high accuracy results. Three types of ne

... Show More
View Publication
Scopus (16)
Crossref (6)
Scopus Crossref