A Dynamic-Static Dual Input Deep Neural Network Algorithm for Diagnosing COVID-19 by Cough

The COVID-19 (corona virus disease 2019) has caused serious impacts worldwide. Many scholars have done a lot of research on the prevention and control of the epidemic. The diagnosis of COVID-19 by cough is non-contact, low-cost, and easy-access, however, such research is still relatively scarce in China. Mel frequency cepstral coefficients (MFCC) feature can only represent the static sound feature, while the first-order differential MFCC feature can also reflect the dynamic feature of sound. In order to better prevent and treat COVID-19, the paper proposes a dynamic-static dual input deep neural network algorithm for diagnosing COVID-19 by cough. Based on Coswara dataset, cough audio is clipped, MFCC and first-order differential MFCC features are extracted, and a dynamic and static feature dual-input neural network model is trained. The model adopts a statistic pooling layer so that different length of MFCC features can be input. The experiment results show the proposed algorithm can significantly improve the recognition accuracy, recall rate, specificity, and F1-score compared with the existing models. © 2023 Chinese Institute of Electronics. All rights reserved.

Face Mask Detection and Recognition with High Accuracy on Live Streaming Video Using Improved Yolo V4 and Comparing with Convolutional Neural Network

The aim of this research is to detect face masks using Convolutional Neural network (CNN) algorithm and comparing it with the Yolo v4 algorithm. The study includes two groups namely, CNN algorithm and yolo v4 algorithm. The total sample size is 40 with pretest power of 0.8. In order to evaluate how well CNN algorithm methods perform, accuracy values are calculated. Using SPSS software, CNN algorithm method was found to be 92.65% accurate while improved Yolo v4 was found to be 85.87% accurate. 0.000 p(2-tailed) is obtained for the model. Using CNN, it was proved significant improvements to performance than improved Yolo v4. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Person Identification and Gender Classification Based on Vision Transformers for Periocular Images

Simple SummaryThe idea of identifying persons using the fewest traits from the face, particularly the area surrounding the eye, was carried out in light of the present COVID-19 scenario. This may also be applied to doctors working in hospitals, the military, and even in certain faiths where the face is mostly covered, except the eyes. The most recent advancement in computer vision, called vision transformers, has been tested for the UBIPr dataset for different architectures. The proposed model is pretrained on an openly available ImageNet dataset with 1 K classes and 1.3 M pictures before using it on the real dataset of interest, and accordingly the input images are scaled to 224 × 224. The PyTorch framework, which is particularly helpful for creating complicated neural networks, has been utilized to create our models. To avoid overfitting, the stratified K-Fold technique is used to make the model less prone to overfitting. The accuracy results have proven that these techniques are highly effective for both person identification and gender classification.AbstractMany biometrics advancements have been widely used for security applications. This field's evolution began with fingerprints and continued with periocular imaging, which has gained popularity due to the pandemic scenario. CNN (convolutional neural networks) has revolutionized the computer vision domain by demonstrating various state-of-the-art results (performance metrics) with the help of deep-learning-based architectures. The latest transformation has happened with the invention of transformers, which are used in NLP (natural language processing) and are presently being adapted for computer vision. In this work, we have implemented five different ViT- (vision transformer) based architectures for person identification and gender classification. The experiment was performed on the ViT architectures and their modified counterparts. In general, the samples selected for train:val:test splits are random, and the trained model may get affected by overfitting. To overcome this, we have performed 5-fold cross-validation-based analysis. The experiment's performance matrix indicates that the proposed method achieved better results for gender classification as well as person identification. We also experimented with train-val-test partitions for benchmarking with existing architectures and observed significant improvements. We utilized the publicly available UBIPr dataset for performing this experimentation.

Improved Accuracy in Speech Recognition System for Detection of COVID-19 using Support Vector Machine and Comparing with Convolution Neural Network Algorithm

The objective of the research aims to detect Covid-19 patients by innovative speech recognition using a Support Vector Machine (SVM) and comparing accuracy with Convolutional Neural Network (CNN). Speech recognition using SVM is considered as group 1 and Convolutional Neural Network is considered as group 2, where each group has 20 samples. A T-test with 95% CI, G-power of 80%, and alpha=0.05 was used to compare the two sets of data. CNN achieves an accuracy of 87.5% and SVM achieves an accuracy of 92.5% with significance value 0.043 (P<0.05). Covid-19 prediction using an innovative speech recognition using SVM achieves significantly better accuracy than CNN. © 2022 IEEE.

Sentiment Analysis of the Use of Telecommunication Providers on Twitter Social Media using Convolutional Neural Network

Telecommunication technology continues to develop starting from 1G, 2G, 3G, 4G, and currently entering the 5G era. The Global System for Mobile Communications (GSM) based telecommunication industry in Indonesia consists of three big names: Telkomsel, XL, and Indosat. During the Covid-19 pandemic, activities carried out outside the home should be done online. People hope that the internet network can work properly. However, the reality is not as expected, because many networks are experiencing slow internet problems and many complaints are delivered through social media. Therefore, this research aims to find the insight opinions that have been conveyed to the telecommunications operator in social media. This research used the Convolutional Neural Network (CNN) algorithm to classify text sentiment (negative or positive) about telecommunication providers. The experiment with text data from Twitter is conducted after preprocessing and weighting of the Word2Vec process. The confusion matrix experiment shows that the CNN algorithm's performance reaches an average accuracy value of around 86.22%. The experiment was carried out by dividing the training data and testing the data 5 times in 10 times. The study results indicated that disruption of cellular telecommunications operators provided many sentiments, especially negative sentiment at the beginning of the COVID-19 pandemic. © 2022 IEEE.

Information Technology for Chest X-Ray Images Analysis

This paper presents the research and development of information technology for analysis and classification of chest X-ray images in order to automatically detect the signs of the disease, specifically pneumonia, what is the most relevant in the conditions of COVID-19 pandemic. Information technology is based on the developed mathematical model through complex training of neural networks. The dataset used for the experimental studies and neural networks training consisted of 35,000 images ranging in size from 200×200 px to 2500×2500 px. Convolutional neural networks were used to fulfill the goal of software creation based on developed information technology. As a result of experiments, the weighted average value of F1 metric of 97.05% was obtained, that is close to the recognition rate of a physician. During the research the decision support software based on developed information technology was created with an aim to assist the physician in making a decision, help in the analysis of lungs X-rays for pneumonia, and also allow to store all the necessary information about the patients in one repository. The program was developed using Microsoft technologies, including the C# programming language and a technology environment designed to develop a user interface - WPF. Also, software was implemented using the MVVM architecture and ML.NET as a tool for implementation of a neural network. The Nvidia RTX 2070 Super graphics processor (GPU) and CUDA technology were used to train the neural network. Created software based on developed information technology for chest X-ray images analysis allows to record patients, classify and process images, add confirmations of physicians, and can be used as an accessory instrument to diagnose pneumonia, which will reduce the strain on the radiologist and allow to process larger number of X-rays images more effective. © 2022 Copyright for this paper by its authors.

Application of Convolutional Neural Network in COVID-19 Diagnosis

Since the outbreak and spread of COVID-19 in large areas of the world, the importance of rapid diagnosis of COVID-19 has increased. In the first week after the onset of COVID-19, the density of lesions is uneven, and chest CT is often difficult to show local subpleural ground-glass shadows, resulting in missed diagnosis. The COVID-19 intelligent diagnosis system based on the convolutional neural network algorithm can not only accurately identify the feature points, reduce the workload of doctors and improve the diagnosis efficiency, but also reduce the rate of missed diagnosis and misdiagnosis, which is conducive to epidemic control. © 2022 ACM.

Development of Deep Neural Network Algorithm for Identification of Cerebral Microstructural Changes in Brain Tumor for Post-COVID-19 Patients

The year 2019 brought the once in hundred years’ experience for the whole world. COVID-19 pandemic shaken almost all segments of everyone’s life and scientists all over the world are engaged in saving our existence. As there is a need of capturing microstructural changes like tumor boundary pixel level shifts and/or growth, deep learning can be a very promising to identify the pixel level changes occurred in brain MR images. The multi-layer execution using CNN architecture is possible, but there is a need for fast convolution and de-convolution with lowered strides. Conventional methods can provide acceptable results, but to identify the microstructural changes in (COVID-19 patient) MR image, accuracy and visibility at pixel level need to be very precise. Hence, this paper presents the methodology for analysis of pre- and post-COVID-19 brain tumor microstructures by means of development of novel CNNPostCoV deep learning algorithm. Proposed research uses IIARD-19 and IIARD-20 dataset of COVID-19 patient. Algorithm framed with convolution neural network architecture which provides better performance of dice score, sensitivity, and PPV parameters. Paper also presents the training and validation analysis for HGG, LGG, and combined dataset of multi-modal brain tumors. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Detection of COVID-19 and Classification of Pneumonia Using Deep Neural Networks Algorithms

COVID-19 which is a subclass of severe acute respiratory syndrome (SARS) is a viral disease which emerged from China in 2019. At first, there are shorthand of test kits available to diagnose the COVID-19 disease. The tests available to diagnose the COVID-19 are RT-PCR (real-time polymerase chain reaction), Rapid Antigen test and Antibody test. But in these, only RT-PCR has the high accuracy, and it is a time-taking process. It takes nearly from 4 to 48 h. Here, AI plays an important role in diagnosing the disease. In the recent years, AI becomes a part of medical field and is widely used in classification. The chest X-Rays are used to detect the COVID-19 using deep learning and the model used to detect the COVID-19 is ResNet18 which is a residual network containing 18 layers. In this work, we classified four types of classes to make sure that our model performance is better and classify accurately. The data set contain a total of 5365 images. In this, we used 80% of data for the training and 20% for validation. The accuracy obtained in classification of three classes is 96.67% and for four classes, the accuracy is 91%. We have also used another model for comparison which ResNet50 and achieved an accuracy of 75%. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Deep Learning Analysis for COVID-19 Using Neural Network Algorithms

The COVID-19 pandemic threatens to devastatingly impact the global population’s safety. A successful surveillance of contaminated patients is a crucial move in the battle against COVID-19, and radiological photographs via chest X-ray are one of the main screening strategies. Recent research showed that patients have abnormalities in photographs of chest X-ray that are characteristic of COVID-19 infects. This has inspired a set of deep learning artificial intelligence (AI) programs, and it has been seen that the precision of the identification of COVID-19 contaminated patients utilizing chest X-rays has been quite positive. However, these built AI schemes, to the extent of their author’s awareness, have become closed sources and not accessible for further learning and expansion by the scientific community, so they are not open to the general public. This thesis therefore implements COVID-Net to identify COVID-19 cases of chest X-rays images, an open source, accessible to the general public, a deep neural network architecture adapted to the detection. The COVID-Net data collection, which is referred to as COVIDx which includes 13,800 chest X-ray photographs of 13,725 patients from 3 open-access data sources, one of which we launched, are also addressed. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Using COVID-19 Masks Dataset to Implement Deep Convolutional Neural Networks For Facial Emotion Recognition

In light of the novel pandemic called COVID-19, the world has been instructed to wear protective facial masks to limit its spread. Doing so has reduced the effectiveness of traditional facial recognition technologies, especially in processing human facial emotions. This has rendered the usage of such technology obsolete in managing facial databases, relying on it for security purposes, and so on. It is then necessary to enhance the current generation of facial recognition to adapt to the protective masks. Speaking of the current facial recognition generation, most of its complex iterations heavily rely on deep learning, which is flawed since the existing facial databases are insufficient, making it even more inadequate to bypass facial masks. This is why the present research paper suggests implementing the Deep Convolutional Neural Networks (DCNN) algorithm using the Japanese Female Facial Expression (JAFFE) to simulate a masked face emotion recognition. This facial database is available free for academic research, was utilized to label the available images displaying various facial emotions under the umbrella of one of the seven basic human facial emotions, allowing for a more advanced facial technology. Consistent with the latest research findings, the proposed facial emotion recognition attains up to an accuracy of 71.35% due to its meticulous masked facial database. © 2021 IEEE.