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 subject of youth care of important issues in view of what constitutes the importance
to the development of societies in general and as much as enjoy young people in any society
are good psychological health and agree psychosocial be healthy to be effective to invest their
energies and their potential for the progress of that society and development of the social
aspects and economic. The universities of the most important educational institutions that
provide care for young people, they are as well as providing information and expertise
necessary to prepare young people for life and the development of mental abilities, they are
different activities that will satisfy their needs physical, psychological, social and
Abstract
The government spending in Iraq and witnessed the changes and developments, especially after 2003, which outweighed consumer spending at the expense of capital expenditure and increased support and diversity of trends towards improving pension conditions for member
... Show MoreABSTRUCT
This research aims at examining the expected gap between the fact of planning and controlling process of production at the State Company for Electric Industries and implementation of material requirements planning system in fuzzy environment. Developing solutions to bridge the gap is required to provide specific mechanisms subject to the logic of fuzzy rules that will keep pace with demand for increased accuracy and reduced waiting times depending on demand forecast, investment in inventory to reduce costs to a minimum.
The proposed solutions for overcoming the research problem has required some questions reflecting the problem with its multiple dimensions, which ar
... Show MoreObjective: The study aimed to determine quality of life domains for adult patients with limbs loss and to identify
the association between quality of life domains and demographic characteristics and medical information.
Methodology: A descriptive study was carried out at Baghdad artificial limb center, Al-Salam medical
rehabilitation center, Al-Ghadeer medical rehabilitation center and the rheumatoid and medical rehabilitation
center for the period from September 2007 to April 2008. A purposive ''non- probability'' sample of (200)
patients with limbs loss. Questionnaire form was constructed for the purpose of the study. Data were collected
through the application of the questionnaire and interview technique. Data were a
Six isolates of A. pullulans were collected from many sources including Hibiscus sabdariffa (Roselle), old Roofs of houses and bathroom surface that referred as Ap ros1, Ap or2, 3, 4 and Ap bs5, 6 respectively, all these isolates were identified based on morphological characteristics and nutritional physiology profiles, all were able to utilize various carbon and nitrogen sources such as glucose, xylose, sucrose, maltose, ammonium sulfate, ammonium nitrate and ammonium chloride, also they showed positive test for starch and amylase, while α-cellulose, ethanol, and methanol were could not be ass
... Show MoreComputer systems and networks are being used in almost every aspect of our daily life; as a result the security threats to computers and networks have also increased significantly. Traditionally, password-based user authentication is widely used to authenticate legitimate user in the current system0T but0T this method has many loop holes such as password sharing, shoulder surfing, brute force attack, dictionary attack, guessing, phishing and many more. The aim of this paper is to enhance the password authentication method by presenting a keystroke dynamics with back propagation neural network as a transparent layer of user authentication. Keystroke Dynamics is one of the famous and inexpensive behavioral biometric technologies, which identi
... Show More