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
Mon Jan 01 2024
Journal Name
Ieee Access
A Magnetic Field Concentration Method for Magnetic Flux Leakage Detection of Rail-Top Surface Cracks
...Show More Authors

View Publication
Scopus Clarivate Crossref
Publication Date
Sun Dec 01 2013
Journal Name
Toxicon
Methods for simultaneous detection of the cyanotoxins BMAA, DABA, and anatoxin-a in environmental samples
...Show More Authors

View Publication Preview PDF
Scopus (33)
Crossref (32)
Scopus Clarivate Crossref
Publication Date
Thu Dec 23 2021
Journal Name
Iraqi Journal Of Science
Design and Fabrication of Nanostructure TiO2 Doped NiO as A Gas Sensor for NO2 Detection
...Show More Authors

      In this paper, thin films of undoped and nickel oxide (NiO) doped titanium dioxide (TiO2) were prepared using the chemical spray pyrolysis deposition (CSP) technique, with different concentrations of nickel oxide (NiO) in the range (3-9) wt%. The morphological, structural, electrical, and sensing properties of a gas of the prepared thin films were examined. XRD measurements showed that TiO2 films have a polycrystalline structure. AFM analysis showed that these films have a regular structure both before and after doping . The roughness of these films  decreased after adding  impurities but then the opposite of that took  place. The electrical and gas sens

... Show More
View Publication Preview PDF
Scopus (2)
Crossref (2)
Scopus Crossref
Publication Date
Sat Dec 17 2022
Journal Name
Iraqi Journal Of Laser
PDF Biosensing technique for detection of H.pylori bacteria: Israa M.L. SaQari*, Layla M.H. Al-ameri
...Show More Authors

Abstract :H.pylori is an important cause of gastric duodenal disease, including gastric ulcers, Mucosa-associated lymphoid tissue (MALT), and gastric carcinoma. biosensors are becoming the most extensively studied discipline because the easy, rapid, low-cost, highly sensitive, and highly selective biosensors contribute to advances in next-generation medicines such as individualized medicine and ultrasensitive point-of-care detection of markers for diseases.  Five of ten patients diagnosed with H.pylori ranging in age from 15–85 participated in this research. who [gastritis, duodenitis, duodenal ulcer (DU), and peptic ulcer (PU)] Suspected H.pylori colonies w

... Show More
View Publication Preview PDF
Publication Date
Wed Dec 30 2020
Journal Name
Iraqi Journal Of Science
Influence of Heat Transfer on MHD Oscillatory Flow for Eyring-Powell Fluid through a Porous Medium with Varying Temperature and Concentration
...Show More Authors

The aim of this research is to study the effect of heat transfer on the oscillating flow of the hydrodynamics magnetizing Eyring-Powell fluid through a porous medium under the influence of temperature and concentration for two types of engineering conditions "Poiseuille flow and Couette flow". We used the perturbation method to obtain a clear formula for fluid motion. The results obtained are illustrated by graphs.

View Publication Preview PDF
Scopus (1)
Scopus Crossref
Publication Date
Sun Aug 01 2021
Journal Name
Journal Of Engineering
Practical comparation of the accuracy and speed of YOLO, SSD and Faster RCNN for drone detection
...Show More Authors

Convolutional Neural Networks (CNN) have high performance in the fields of object recognition and classification. The strength of CNNs comes from the fact that they are able to extract information from raw-pixel content and learn features automatically. Feature extraction and classification algorithms can be either hand-crafted or Deep Learning (DL) based. DL detection approaches can be either two stages (region proposal approaches) detector or a single stage (non-region proposal approach) detector. Region proposal-based techniques include R-CNN, Fast RCNN, and Faster RCNN. Non-region proposal-based techniques include Single Shot Detector (SSD) and You Only Look Once (YOLO). We are going to compare the speed and accuracy of Faster RCNN,

... Show More
View Publication Preview PDF
Crossref (13)
Crossref
Publication Date
Sun Mar 07 2010
Journal Name
Baghdad Science Journal
Detection of BRCA1and BRCA2 mutation for Breast Cancer in Sample of Iraqi Women above 40 Years
...Show More Authors

Breast cancer is the commonest cancer affecting women worldwide. Different studies have dealt with the etiological factors of that cancer aiming to find a way for early diagnosis and satisfactory therapy. The present study clarified the relationship between genetic polymorphisms of BRCA1 & BRCA2 genes and some etiological risk factors among breast cancer patients in Iraq. This investigation was carried out on 25 patients (all were females) who were diagnosed as breast cancer patients attended AL-Kadhemya Teaching Hospital in Baghdad and 10 apparently healthy women were used as a control, all women (patients and control) aged above 40 years. The Wizard Promega kit was used for DNA isolation from breast patients and normal individuals. B

... Show More
View Publication Preview PDF
Crossref
Publication Date
Mon Jan 01 2024
Journal Name
Lecture Notes In Electrical Engineering
A Study of the Influence of Steel Brushes in Rail Surface Magnetic Flux Leakage Detection Using Finite Elements Simulation
...Show More Authors

View Publication
Scopus Crossref
Publication Date
Fri Nov 24 2023
Journal Name
Iraqi Journal Of Science
Isolation and Identification of Fungal Propagation in Stored Maize and detection of aflatoxin B1 Using TLC and ELISA Technique
...Show More Authors

Aflatoxin B1 (AFB1) is a mycotoxin produced mainly by fungi Aspergillus flavus in food and animals feed. It is considered as a carcinogenic toxin for human and animals. The current study is designed to investigate the incidence of mycoflora in twenty four samples of local stored maize collected from Iraqi governorates; investigate the presence of aflatoxin B1 on these samples using TLC and ELISA techniques. The fungi recovered from maize samples were Aspergillus flavus (18.57 % ), Fusarium spp. (12.8 % ), A. ocraceus (9.96 % ) , A. terrus (9.07 % ), A. fumigatus (8.46 % ) , Alternaria spp. (6.40 % ) Rhizopus spp. (4.98 % ), A. niger spp., A. oryzae spp. (4.80 % ), Penicillium spp. (4.53 %) A. versicolor spp., Rhizoctonia spp. (4.27 %), A

... Show More
View Publication Preview PDF
Publication Date
Mon Sep 30 2002
Journal Name
Iraqi Journal Of Chemical And Petroleum Engineering
Volumetric Mass Transfer Capacity Coefficient in the RTL Extractor
...Show More Authors

View Publication Preview PDF