An Effective Deep Learning Model for Health Monitoring and Detection of COVID-19 Infected Patients: An End-to-End Solution.

The COVID-19 infection is the greatest danger to humankind right now because of the devastation it causes to the lives of its victims. It is important that infected people be tested in a timely manner in order to halt the spread of the disease. Physical approaches are time-consuming, expensive, and tedious. As a result, there is a pressing need for a cost-effective and efficient automated tool. A convolutional neural network is presented in this paper for analysing X-ray pictures of patients' chests. For the analysis of COVID-19 infections, this study investigates the most suitable pretrained deep learning models, which can be integrated with mobile or online apps and support the mobility of diagnostic instruments in the form of a portable tool. Patients can use the smartphone app to find the nearest healthcare testing facility, book an appointment, and get instantaneous results, while healthcare professionals can keep track of the details thanks to the web and mobile applications built for this study. Medical practitioners can apply the COVID-19 detection model for chest frontal X-ray pictures with ease. A user-friendly interface is created to make our end-to-end solution paradigm work. Based on the data, it appears that the model could be useful in the real world.

Classification of Transgenic Mice by Retinal Imaging Using SVMS.

Alzheimer's disease is the neuro disorder which characterized by means of Amyloid- ß (A ß) in brain. However, accurate detection of this disease is a challenging task since the pathological issues of brain are complex in identification. In this paper, the changes associated with the retinal imaging for Alzheimer's disease are classified into two classes such as wild-type (WT) and transgenic mice model (TMM). For testing, optical coherence tomography (OCT) images are used to classify into two groups. The classification is implemented by support vector machines with the optimum kernel selection using a genetic algorithm. Among several kernel functions of SVM, the radial basis kernel function provides the better classification result. In order to deal with an effective classification using SVM, texture features of retinal images are extracted and selected. The overall accuracy reached 92% and 91% of precision for the classification of transgenic mice.

Cloud-Based IoE Enabled an Urban Flooding Surveillance System.

A flood is defined as a surplus of water or sludge on parched soil, and a flood has originated through the runoff of water inside the water route from the various water sources like canals, etc. Intense rainfall, deforestation, urbanization, deprived water and sewerage administration, and lack of concentration toward the environment of the hydrological scheme have been the causes of urban flooding. In addition, there is a deficiency in flood assessment due to the impediment in getting data on floods to the control room from the flood-affected area. To diminish the possessions due to flooding, there ought to be an immediate move of captured statistics as of the hectic region en route to the observation room with no further wait for a completely fledged technique in the wireless settings data from the Internet of Things (IoT). The Internet of Everything (IoE) is a concept that extends the Internet of Things. In view of the fact that the wireless nodes are changeable in their environment, those effects lead to unsteadiness and uncertainty in information distribution. Therefore, there is a requirement for flood-predictable region data that may be exaggerated between the source and the control room. In the past, there were a lot of techniques set up and put into practice intended for keeping an eye on the flood spots. However, one of the biggest challenges is to have data sharing without delay and loss of data among source and destination nodes. In addition to that, the video quality also needs to be taken into consideration at the same time in receipt, as it is a tough task to determine and preplan the flood happenings completely from the normal disaster that makes scientific complicatedness more than the information being received in a wireless ad-hoc environment using IoT-based sensors. Considering all the abovementioned reasons, the proposed work comprises of three folded goals, namely, the design of a mobile ad-hoc flooding environment, the development of an urban flood high definition video surveillance system using IoT-based sensors, and experimental work on simulation.

Human-Computer Interaction with Hand Gesture Recognition Using ResNet and MobileNet.

Sign language is the native language of deaf people, which they use in their daily life, and it facilitates the communication process between deaf people. The problem faced by deaf people is targeted using sign language technique. Sign language refers to the use of the arms and hands to communicate, particularly among those who are deaf. This varies depending on the person and the location from which they come. As a result, there is no standardization about the sign language to be used; for example, American, British, Chinese, and Arab sign languages are all distinct. Here, in this study we trained a model, which will be able to classify the Arabic sign language, which consists of 32 Arabic alphabet sign classes. In images, sign language is detected through the pose of the hand. In this study, we proposed a framework, which consists of two CNN models, and each of them is individually trained on the training set. The final predictions of the two models were ensembled to achieve higher results. The dataset used in this study is released in 2019 and is called as ArSL2018. It is launched at the Prince Mohammad Bin Fahd University, Al Khobar, Saudi Arabia. The main contribution in this study is resizing the images to 64 ∗ 64 pixels, converting from grayscale images to three-channel images, and then applying the median filter to the images, which acts as lowpass filtering in order to smooth the images and reduce noise and to make the model more robust to avoid overfitting. Then, the preprocessed image is fed into two different models, which are ResNet50 and MobileNetV2. ResNet50 and MobileNetV2 architectures were implemented together. The results we achieved on the test set for the whole data are with an accuracy of about 97% after applying many preprocessing techniques and different hyperparameters for each model, and also different data augmentation techniques.

Comparison of different optimizers implemented on the deep learning architectures for COVID-19 classification.

COVID-19 is the present-day pandemic around the globe. WHO has estimated that approx 15% of the world's population may have been infected with coronavirus with a large number of population on the verge of being infected. It is quite difficult to break the virus chain since asymptomatic patients can result in the spreading of the infection apart from the seriously infected patients. COVID-19 has many similar symptoms to SARS-D however, the symptoms can worsen depending on the immunity power of the patients. It is necessary to be able to find the infected patients even with no symptoms to be able to break the spread of the chain. In this paper, the comparison table describes the accuracy of deep learning architectures by the implementation of different optimizers with different learning rates. In order to remove the overfitting issue, different learning rate has been experimented. Further in this paper, we have proposed the classification of the COVID-19 images using the ensemble of 2 layered Convolutional Neural Network with the Transfer learning method which consumed lesser time for classification and attained an accuracy of nearly 90.45%.

Evaluation of the Protection Provided by Hepatitis B Vaccination in India.

OBJECTIVE: In India, Hepatitis B vaccination is recommended at 6 wk except for hospital-deliveries. The authors examined protection afforded by the birth dose. METHODS: A case-control study was done. HBsAg and HBcAb were tested in 2671 children, 1 to 5 y and HBsAb was evaluated in a subset of 1413 children. Vaccination history was recorded. Cases were HBsAg carriers. In another analysis, children who got infected (HBsAg and/or HBcAb positive) were considered as cases. Exposed were the unvaccinated. In another analysis, exposed were those vaccinated without the birth dose. RESULTS: The odds ratio (OR) for HBsAg positivity with birth vaccination was 0.35 (95% CI 0.19-0.66); while with vaccination at 6 wk was 0.29 (95%CI 0.14-0.61), both compared to unvaccinated. Birth vaccination has no added protection when compared to the unvaccinated. Unvaccinated children in index study had HBsAg positivity of 4.38%. The number needed to treat (NNT) to prevent one case of HBsAg positivity was 32.6 (95% CI, 20.9 to 73.6). The odds of getting HBV infection was 0.42 (CI 0.25-0.68) with birth dose and 0.49 (CI 0.30-0.82) without the birth dose compared to the unvaccinated. Protective antibody (HBsAb) was present in about 70% of the vaccinated. In the unimmunised, in the first 2 y HBsAb protection was present in 40%. The odds ratio (OR) for HBsAb in the fully vaccinated between 4 and 5 y was 1.4 (95%CI 0.9-2.18) compared to the unvaccinated. CONCLUSIONS: The present study lends support to the pragmatic approach of the Government to vaccinate babies born at home starting at 6 wk.