Detection of COVID-19 from chest X-ray images: Boosting the performance with convolutional neural network and transfer learning.

Coronavirus disease (COVID-19) is a pandemic that has caused thousands of casualties and impacts all over the world. Most countries are facing a shortage of COVID-19 test kits in hospitals due to the daily increase in the number of cases. Early detection of COVID-19 can protect people from severe infection. Unfortunately, COVID-19 can be misdiagnosed as pneumonia or other illness and can lead to patient death. Therefore, in order to avoid the spread of COVID-19 among the population, it is necessary to implement an automated early diagnostic system as a rapid alternative diagnostic system. Several researchers have done very well in detecting COVID-19; however, most of them have lower accuracy and overfitting issues that make early screening of COVID-19 difficult. Transfer learning is the most successful technique to solve this problem with higher accuracy. In this paper, we studied the feasibility of applying transfer learning and added our own classifier to automatically classify COVID-19 because transfer learning is very suitable for medical imaging due to the limited availability of data. In this work, we proposed a CNN model based on deep transfer learning technique using six different pre-trained architectures, including VGG16, DenseNet201, MobileNetV2, ResNet50, Xception, and EfficientNetB0. A total of 3886 chest X-rays (1200 cases of COVID-19, 1341 healthy and 1345 cases of viral pneumonia) were used to study the effectiveness of the proposed CNN model. A comparative analysis of the proposed CNN models using three classes of chest X-ray datasets was carried out in order to find the most suitable model. Experimental results show that the proposed CNN model based on VGG16 was able to accurately diagnose COVID-19 patients with 97.84% accuracy, 97.90% precision, 97.89% sensitivity, and 97.89% of F1-score. Evaluation of the test data shows that the proposed model produces the highest accuracy among CNNs and seems to be the most suitable choice for COVID-19 classification. We believe that in this pandemic situation, this model will support healthcare professionals in improving patient screening.

Deep Residual Network for Diagnosis of Retinal Diseases Using Optical Coherence Tomography Images.

Diabetic retinopathy occurs due to damage to the blood vessels in the retina, and it is a major health problem in recent years that progresses slowly without recognizable symptoms. Optical coherence tomography (OCT) is a popular and widely used noninvasive imaging modality for the diagnosis of diabetic retinopathy. Accurate and early diagnosis of this disease using OCT images is crucial for the prevention of blindness. In recent years, several deep learning methods have been very successful in automating the process of detecting retinal diseases from OCT images. However, most methods face reliability and interpretability issues. In this study, we propose a deep residual network for the classification of four classes of retinal diseases, namely diabetic macular edema (DME), choroidal neovascularization (CNV), DRUSEN and NORMAL in OCT images. The proposed model is based on the popular architecture called ResNet50, which eliminates the vanishing gradient problem and is pre-trained on large dataset such as ImageNet and trained end-to-end on the publicly available OCT image dataset. We removed the fully connected layer of ResNet50 and placed our new fully connected block on top to improve the classification accuracy and avoid overfitting in the proposed model. The proposed model was trained and evaluated using different performance metrics, including receiver operating characteristic (ROC) curve on a dataset of 84,452 OCT images with expert disease grading as DRUSEN, CNV, DME and NORMAL. The proposed model provides an improved overall classification accuracy of 99.48% with only 5 misclassifications out of 968 test samples and outperforms existing methods on the same dataset. The results show that the proposed model is well suited for the diagnosis of retinal diseases in ophthalmology clinics.

Double cascade dressed MOSFET from doped Eu3+and Pr3+ in a host YPO4.

In this paper, we study double cascade dressed optical metal oxide semiconductor field-effect transistor (MOSFET) by exploiting enhancement and suppression for mixed-phase (hexagonal + tetragonal) of Eu3+:YPO4 and different phases (hexagonal + tetragonal and pure tetragonal) of Pr3+:YPO4 crystals. We report variation of fine structure energy levels in different doped ions (Eu3+ and Pr3+) in the host YPO crystal. We compared multi-level energy transition from a single dressing laser with single level energy transition from double cascade dressing lasers. Gate delay facilitates multi-energy level dressed transition and is modeled through a Hamiltonian. Based on the results of double cascade dressing, we have realized MOSFET for logic gates (inverter and logic not and gate) with a switching contrast of about 92% using a mixed phase of Pr3+:YPO4.