I remember when I was a child, my grandfather used to talk about outbreaks before I was born. I never liked those stories despite his good intention to entertain me :) I never thought that I would experience it at some point in my lifetime. He passed away 5 years ago, well before this pandemic. It seems he wanted to open my eyes and prepare me for a day like today; who knows! Thanks grandpa for your never-ending wisdom in my life even when you are no longer with us.
COVID-19 pandemic has impacted all of us sadly. As someone who has lost family members to this disease, I can feel the pain and sorrow more than anybody else. But why am I sharing this in my website and creating this small page. Well, to do what I feel it is right to do. Research changes lives for good. I now felt the positive impact that research can bring to the world more then ever. As a small researcher in the area of AI, I decided to involve in and initiate several projects as my contribution to handle this disruptive event. I feel very good to be a tiny tiny part of the global move agaist this unprecedented event in the century.
Coronavirus disease 2019 (COVID-19) is a contagious respiratory and vascular disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a specific type of coronavirus, responsible for an ongoing pandemic. Common symptoms include fever, cough, fatigue, breathing difficulties, and loss of smell and taste. Symptoms begin one to fourteen days after exposure to the virus. While most people have mild symptoms, some people develop acute respiratory distress syndrome (ARDS) possibly precipitated by cytokine storm, multi-organ failure, septic shock, and blood clots.
During a crisis such as COVID-19, the demand for information drastically increases. The term infodemic has been even used by WHO, which indicates the fact that the risks of fake news are as high as the health risks of COVID-19. In this project, we are harnessing the power of AI in detecting fake news. To cope with these challenges, we are designing AI frameworks and tools that can be used at three different stage: pre-moderation, post-moderation, and reactive-moderation.
• As a pre-moderator, AI can be trained with a large-scale dataset of news to classify news as fake or true. As discussed above, this classification is done after training is done. If AI detect news that cannot be classified, it can flag it for human moderation.
• As a post-moderator, AI can be used to synthesizes data for the training dataset to increase the accuracy of AI pre-moderator. It can also be used to process and analyze flagged content by users after publication too.
• As a reactive moderator, AI can assist the stage of moderation at requires human intervention. In other words, AI can facilitate and speed up the process of human content moderation. For instance, key frames in a video news can be highlighted to be moderated by a human rather than watching the entire news.
This paper aims to classify COVID-19, normal, and pneumonia patients from chest X-ray images. As such, an Optimized Convolutional Neural network (OptCoNet) is proposed in this work for the automatic diagnosis of COVID-19. The proposed OptCoNet architecture is composed of optimized feature extraction and classification components. The Grey Wolf Optimizer (GWO) algorithm is used to optimize the hyperparameters for training the CNN layers. The proposed model is tested and compared with different classification strategies utilizing an openly accessible dataset of COVID-19, normal, and pneumonia images. The presented optimized CNN model provides accuracy, sensitivity, specificity, precision, and F1 score values of 97.78%, 97.75%, 96.25%, 92.88%, and 95.25%, respectively, which are better than those of state-of-the-art models. This proposed CNN model can help in the automatic screening of COVID-19 patients and decrease the burden on medicinal services frameworks.
This article proposes two optimization algorithms for feature selection and classification of COVID-19. The proposed framework has three cascaded phases. Firstly, the features are extracted from the CT scans using a Convolutional Neural Network (CNN) named AlexNet. Secondly, a proposed features selection algorithm, Guided Whale Optimization Algorithm (Guided WOA) based on Stochastic Fractal Search (SFS), is then applied followed by balancing the selected features. Finally, a proposed voting classifier, Guided WOA based on Particle Swarm Optimization (PSO), aggregates different classifiers’ predictions to choose the most voted class. This increases the chance that individual classifiers, e.g. Support Vector Machine (SVM), Neural Networks (NN), k-Nearest Neighbor (KNN), and Decision Trees (DT), to show significant discrepancies. Two datasets are used to test the proposed model: CT images containing clinical findings of positive COVID-19 and CT images negative COVID-19. The proposed feature selection algorithm (SFS-Guided WOA) is compared with other optimization algorithms widely used in recent literature to validate its efficiency. The proposed voting classifier (PSO-Guided-WOA) achieved AUC (area under the curve) of 0.995 that is superior to other voting classifiers in terms of performance metrics. Wilcoxon rank-sum, ANOVA, and T-test statistical tests are applied to statistically assess the quality of the proposed algorithms as well.
The rapid expansion of information science has caused the issue of “the curse of dimensionality”, which will negatively affect the performance of the machine learning model. Feature selection is typically considered as a pre-processing mechanism to find an optimal subset of features from a given set of all features in the data mining process. In this project, a novel Hyper Learning Binary Dragonfly Algorithm (HLBDA) is proposed as a wrapper-based method to find an optimal subset of features for a given classification problem. The DA is one of my widely-used optimization algorithms. The proposed method is applied to a coronavirus disease (COVID-19) datasets. The results demonstrate the superiority of HLBDA in increasing prediction accuracy and reducing the number of selected features