Time series prediction of COVID-19 by mutation rate analysis using recurrent neural network-based LSTM model.

SARS-CoV-2, a novel coronavirus mostly known as COVID-19 has created a global pandemic. The world is now immobilized by this infectious RNA virus. As of June 15, already more than 7.9 million people have been infected and 432k people died. This RNA virus has the ability to do the mutation in the human body. Accurate determination of mutation rates is essential to comprehend the evolution of this virus and to determine the risk of emergent infectious disease. This study explores the mutation rate of the whole genomic sequence gathered from the patient's dataset of different countries. The collected dataset is processed to determine the nucleotide mutation and codon mutation separately. Furthermore, based on the size of the dataset, the determined mutation rate is categorized for four different regions: China, Australia, the United States, and the rest of the World. It has been found that a huge amount of Thymine (T) and Adenine (A) are mutated to other nucleotides for all regions, but codons are not frequently mutating like nucleotides. A recurrent neural network-based Long Short Term Memory (LSTM) model has been applied to predict the future mutation rate of this virus. The LSTM model gives Root Mean Square Error (RMSE) of 0.06 in testing and 0.04 in training, which is an optimized value. Using this train and testing process, the nucleotide mutation rate of 400th patient in future time has been predicted. About 0.1% increment in mutation rate is found for mutating of nucleotides from T to C and G, C to G and G to T. While a decrement of 0.1% is seen for mutating of T to A, and A to C. It is found that this model can be used to predict day basis mutation rates if more patient data is available in updated time.

Development of a Robust Multi-Scale Featured Local Binary Pattern for Improved Facial Expression Recognition.

Compelling facial expression recognition (FER) processes have been utilized in very successful fields like computer vision, robotics, artificial intelligence, and dynamic texture recognition. However, the FER's critical problem with traditional local binary pattern (LBP) is the loss of neighboring pixels related to different scales that can affect the texture of facial images. To overcome such limitations, this study describes a new extended LBP method to extract feature vectors from images, detecting each image from facial expressions. The proposed method is based on the bitwise AND operation of two rotational kernels applied on LBP(8,1) and LBP(8,2) and utilizes two accessible datasets. Firstly, the facial parts are detected and the essential components of a face are observed, such as eyes, nose, and lips. The portion of the face is then cropped to reduce the dimensions and an unsharp masking kernel is applied to sharpen the image. The filtered images then go through the feature extraction method and wait for the classification process. Four machine learning classifiers were used to verify the proposed method. This study shows that the proposed multi-scale featured local binary pattern (MSFLBP), together with Support Vector Machine (SVM), outperformed the recent LBP-based state-of-the-art approaches resulting in an accuracy of 99.12% for the Extended Cohn-Kanade (CK+) dataset and 89.08% for the Karolinska Directed Emotional Faces (KDEF) dataset.

Sign language recognition using the fusion of image and hand landmarks through multi-headed convolutional neural network.

Sign Language Recognition is a breakthrough for communication among deaf-mute society and has been a critical research topic for years. Although some of the previous studies have successfully recognized sign language, it requires many costly instruments including sensors, devices, and high-end processing power. However, such drawbacks can be easily overcome by employing artificial intelligence-based techniques. Since, in this modern era of advanced mobile technology, using a camera to take video or images is much easier, this study demonstrates a cost-effective technique to detect American Sign Language (ASL) using an image dataset. Here, "Finger Spelling, A" dataset has been used, with 24 letters (except j and z as they contain motion). The main reason for using this dataset is that these images have a complex background with different environments and scene colors. Two layers of image processing have been used: in the first layer, images are processed as a whole for training, and in the second layer, the hand landmarks are extracted. A multi-headed convolutional neural network (CNN) model has been proposed and tested with 30% of the dataset to train these two layers. To avoid the overfitting problem, data augmentation and dynamic learning rate reduction have been used. With the proposed model, 98.981% test accuracy has been achieved. It is expected that this study may help to develop an efficient human-machine communication system for a deaf-mute society.