U-Net based Semantic Segmentation for Touchless Fingerprint Technology: A Reflective Review

Touch-based fingerprints are widely used in today's world;even with all the success, the touch-based nature of these is a threat, especially in this COVID-19 period. A solution to the same is the introduction of Touchless Fingerprint Technology. The workflow of a touchless system varies vastly from its touch-based counterpart in terms of acquisition, pre-processing, image enhancement, and fingerprint verification. One significant difference is the methods used to segment desired fingerprint regions. This literature focuses on pixel-level classification or semantic segmentation using U-Net, a key yet challenging task. A plethora of semantic segmentation methods have been applied in this field. In this literature, a spectrum of efforts in the field of semantic segmentation using U-Net is investigated along with the components that are integral while training and testing a model, like optimizers, loss functions, and metrics used for evaluation and enumeration of results obtained. © 2022 IEEE.

Covid-19 Detection and Classification using Transfer learning with XGboost

The 2019 new corona virus (COVID-19), with a genesis phase in China, has dispersed apace amid individuals subsisting in distinct nations and is rising toward about twelve lakh cases in the balance as per the intuition of the European center for Health Security and Communicable diseases and ECDC. There is a foreordained figure of COVID-19 trial caskets attainable in medical centers because of the escalating cases in day-to-day life. In this way, it is important to execute a programmed location framework as a snappy elective conclusion alternative to forestall COVID-19 transmitting between peoples. In this examination, three disparate Convolutional neural system- based models (XGBOOST/LIGHTGBM, Inception-ResNetV2 and InceptionV3) have been put forward for the whereabouts of coronavirus and pneumonia contaminated convalescent by harnessing thoracic radiographic screening. Receiver Operating Characteristics (ROC) investigations and disordered networks by those tripartite models are bestowed and deteriorated by exploiting 5-superimpose traverse accredit. Contemplating the demonstration outcome obtained, it is perceived that the pre- prepared XGBOOST/LIGHTGBM model accouters the most upraised characterization execution with 98.6% exactness amongst the other two propounded models (96% correctness for InceptionV3 and 85% exactness for Inception-ResNetV2). © 2023 IEEE.

Ensemble deep honey architecture for COVID-19 prediction using CT scan and chest X-ray images.

Recently, the infectious disease COVID-19 remains to have a catastrophic effect on the lives of human beings all over the world. To combat this deadliest disease, it is essential to screen the affected people quickly and least inexpensively. Radiological examination is considered the most feasible step toward attaining this objective; however, chest X-ray (CXR) and computed tomography (CT) are the most easily accessible and inexpensive options. This paper proposes a novel ensemble deep learning-based solution to predict the COVID-19-positive patients using CXR and CT images. The main aim of the proposed model is to provide an effective COVID-19 prediction model with a robust diagnosis and increase the prediction performance. Initially, pre-processing, like image resizing and noise removal, is employed using image scaling and median filtering techniques to enhance the input data for further processing. Various data augmentation styles, such as flipping and rotation, are applied to capable the model to learn the variations during training and attain better results on a small dataset. Finally, a new ensemble deep honey architecture (EDHA) model is introduced to effectively classify the COVID-19-positive and -negative cases. EDHA combines three pre-trained architectures like ShuffleNet, SqueezeNet, and DenseNet-201, to detect the class value. Moreover, a new optimization algorithm, the honey badger algorithm (HBA), is adapted in EDHA to determine the best values for the hyper-parameters of the proposed model. The proposed EDHA is implemented in the Python platform and evaluates the performance in terms of accuracy, sensitivity, specificity, precision, f1-score, AUC, and MCC. The proposed model has utilized the publicly available CXR and CT datasets to test the solution's efficiency. As a result, the simulated outcomes showed that the proposed EDHA had achieved better performance than the existing techniques in terms of Accuracy, Sensitivity, Specificity, Precision, F1-Score, MCC, AUC, and Computation time are 99.1%, 99%, 98.6%, 99.6%, 98.9%, 99.2%, 0.98, and 820 s using the CXR dataset.

Diffusion Convolutional Recurrent Neural Network-Based Load Forecasting During COVID-19 Pandemic Situation

Infected by the novel coronavirus (COVID-19 – C-19) pandemic, worldwide energy generation and utilization have altered immensely. It remains unfamiliar in any case that traditional short-term load forecasting methodologies centered upon single-task, single-area, and standard signals could precisely catch the load pattern during the C-19 and must be cautiously analyzed. An effectual administration and finer planning by the power concerns remain of higher importance for precise electrical load forecasting. There presents a higher degree of unpredictability's in the load time series (TS) that remains arduous in doing the precise short-term load forecast (SLF), medium-term load forecast (MLF), and long-term load forecast (LLF). For excerpting the local trends and capturing similar patterns of short and medium forecasting TS, we proffer Diffusion Convolutional Recurrent Neural Network (DCRNN), which attains finer execution and normalization by employing knowledge transition betwixt disparate forecasting jobs. This as well evens the portrayals if many layers remain stacked. The paradigms have been tested centered upon the actual life by performing comprehensive experimentations for authenticating their steadiness and applicability. The execution has been computed concerning squared error, Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Error (MAE). Consequently, the proffered DCRNN attains 0.0534 of MSE in the Chicago area, 0.1691 of MAPE in the Seattle area, and 0.0634 of MAE in the Seattle area. © 2022 Lavoisier. All rights reserved.

Predicting ICU Admissions based on Preconditions and Comorbidity for SARS COV2 Infected Patients

Hospitals across the globe have severe constraints in regard to ICU facilities, beds, and other life support systems. However, in certain situations including natural calamities, epidemics or pandemics, large-scale accidents, and so on, the requirement for ICU beds and resources immediately gets augmented. During such times, there exists an impending need for an optimum apportioning of ICU admissions and resources so that those patients who need critical care are given at the right point of time. The onslaught of COVID-19 pandemic has exuded a high probability of virus transmissions and subsequent complications in patients with co-morbidities and relevant medical issues, resulting in the exploration and investigation of models that could forecast the need for ICU admissions with a higher degree of accuracy. In this research study, a patient's pre-condition dataset will be used that is categorical in nature. Feature selection and extractions are implemented and the modified descriptors are provided as input to the model, for evaluating them based on the metrics namely F1-score, accuracy, specificity, and sensitivity. The prime objective is to build a predictive algorithm that will predict prior to the necessity of ICU admissions based on the patient's comorbidity/ precondition specifically for SARS COV2 infection. © 2022 IEEE.

Engagement Estimation using Time-series Facial and Body Features in an Unstable Dataset

The global education sector has been deeply shaken by COVID-19 and forced to shift to an online teaching model. However, the lack of face-to-face communication and interaction in online learning is critical to high-quality teaching and learning. Research on engagement is a crucial part of solving this problem. Because engagement is of time-series data with an ongoing change, research datasets used for engagement analysis need a certain preprocessing method to capture time-series related engagement features. This research proposed a novel deep learning preprocessing method for improving engagement estimation using time-series facial and body information to restore traditional scenes in online learning environments. Such information includes head pose, mouth shape, eye movement, and body distance from the screen. We conducted a preliminary experiment on the DAiSEE dataset for engagement estimation. We applied skipped moving average in data preprocessing to reduce the influence of the extracted noises and oversampled the low engagement level data to balance the engaged/unengaged data. Since engagement is continuous and cannot be captured at a particular instant in time or single images, temporal video classification generally performs better than static classifiers. Therefore, we adopted long short-term memory (LSTM) and Quasi-recurrent neural networks (QRNNs)sequence models to train models and achieved the correct rate of 55.7% (LSTM) and 51.1% (QRNN) using the original key points extracted from OpenPose. Finally, we proposed the optimization structure network achieved the engagement estimation correct rate of 68.5% in proposed LSTM models and 64.2% in QRNN models. The achieved correct rate is 10% higher than the baseline in the DAiSEE dataset. © 30th International Conference on Computers in Education Conference, ICCE 2022 - Proceedings.

Public Opinion Analysis for the Covid-19 Pandemic Based on Sina Weibo Data

The outbreak of Covid-19 has been continuously affecting human lives and communities around the world in many ways. In order to effectively prevent and control the Covid-19 pandemic, public opinion is analyzed based on Sina Weibo data in this paper. Firstly the Weibo data was crawled from Sina website to be the experimental dataset. After preprocessing operations of data cleaning, word segmentation and stop words removal, Term Frequency Inverse Document Frequency (TF-IDF) method was used to perform feature extraction and vectorization. Then public opinion for the Covid-19 pandemic was analyzed, which included word cloud analysis based on text visualization, topic mining based on Latent Dirichlet Allocation (LDA) and sentiment analysis based on Naïve Bayes. The experimental results show that public opinion analysis based on Sina Weibo data can provide effective data support for prevention and control of the Covid-19 pandemic. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

FADCOVNET: Fast Automatic Detection Covid-19 Based on Inception-Resnet-V2 Model

Covid-19 is still a threat to human health. Initial handling in detecting the status of positive COVID-19 patients or not through the IT sector is still very much needed to help the government control the covid-19 outbreak. This study offers a new framework of deep learning classification to help radiologists work in auto-detecting cases of COVID-19 by processing patient X-Ray chest (we call it FADCOVNET). By combining pre-processing techniques with a modified Inception Resnet V2 trained network on the Fully Connected layer and by adding pre-processing data. To control overfitting, the data augmentation method is used. The FADCOVNET model will be compared with the transfer learning model (Resnet50, Inception V3, Inception-Resnet-V3).The dataset used in this study is chest X-ray data for COVID cases as many as 4369 total data. In addition, this study also tested the performance of FADCOVNET on the Covid and healthy chest CT-Scan dataset of 8467 total data. The test results show that the performance of FADCOVNET on the accuracy, sensitivity, specification, precision, and F1-Score are 97%, 98%, 97%, 95%, and 96%, respectively. The results obtained outperform other transfer models. while the accuracy obtained from testing with the CT Scan dataset is 97%. This proves that the FADCOVNET model that we have built can ensure the generalizability of the model very well. From this test, it can be concluded that the proposed CNN architecture works very well in detecting COVID-19. © 2022 IEEE.

CovFrameNet: An Enhanced Deep Learning Framework for COVID-19 Detection.

The novel coronavirus, also known as COVID-19, is a pandemic that has weighed heavily on the socio-economic affairs of the world. Research into the production of relevant vaccines is progressively being advanced with the development of the Pfizer and BioNTech, AstraZeneca, Moderna, Sputnik V, Janssen, Sinopharm, Valneva, Novavax and Sanofi Pasteur vaccines. There is, however, a need for a computational intelligence solution approach to mediate the process of facilitating quick detection of the disease. Different computational intelligence methods, which comprise natural language processing, knowledge engineering, and deep learning, have been proposed in the literature to tackle the spread of coronavirus disease. More so, the application of deep learning models have demonstrated an impressive performance compared to other methods. This paper aims to advance the application of deep learning and image pre-processing techniques to characterise and detect novel coronavirus infection. Furthermore, the study proposes a framework named CovFrameNet., which consist of a pipelined image pre-processing method and a deep learning model for feature extraction, classification, and performance measurement. The novelty of this study lies in the design of a CNN architecture that incorporates an enhanced image pre-processing mechanism. The National Institutes of Health (NIH) Chest X-Ray dataset and COVID-19 Radiography database were used to evaluate and validate the effectiveness of the proposed deep learning model. Results obtained revealed that the proposed model achieved an accuracy of 0.1, recall/precision of 0.85, F-measure of 0.9, and specificity of 1.0. Thus, the study's outcome showed that a CNN-based method with image pre-processing capability could be adopted for the pre-screening of suspected COVID-19 cases, and the confirmation of RT-PCR-based detected cases of COVID-19.

Covid-19 classification using sigmoid based hyper-parameter modified DNN for CT scans and chest X-rays.

Coronavirus disease (COVID-19) is an infectious disease caused by the SARS-CoV-2 virus. Diagnosis of Computed Tomography (CT), and Chest X-rays (CXR) contains the problem of overfitting, earlier diagnosis, and mode collapse. In this work, we predict the classification of the Corona in CT and CXR images. Initially, the images of the dataset are pre-processed using the function of an adaptive Gaussian filter for de-nosing the image. Once the image is pre-processed it goes to Sigmoid Based Hyper-Parameter Modified DNN(SHMDNN). The hyperparameter modification makes use of the optimization algorithm of adaptive grey wolf optimization (AGWO). Finally, classification takes place and classifies the CT and CXR images into 3 categories namely normal, Pneumonia, and COVID-19 images. Better accuracy of 99.9% is reached when compared to different DNN networks.

E-learning course recommendation based on sentiment analysis using hybrid Elman similarity

Online learning is also referred to as E-learning which has gained huge attention and attracted most people during the COVID-19 lockdowns. Due to the excess of online information, users face severe challenges and difficulties realizing the best course that is being competitive in the global market. Therefore, it is necessary to develop an online recommendation system that supports the users in selecting the finest course with E-learning. Thus, the proposed work develops a robust RS model using different approaches. Initially, the pre-processing stage is performed to reduce the presented noise in the website data. Then, the feature extraction stage is done to extract the needed features using Improved TF-IDF, W2V (Word 2 Vector), and Hybrid N-gram. Finally, Elman Minimal Redundancy Maximum Relevance and Enhanced Aquila Optimization (EMRMR_EAO) model is proposed to provide Robust course recommendations. In this work, the ERNN method is used to classify the sentiments based on the similarity measure of the MRMR model. The top course recommendation is afforded depending on the similarity scores like Jaccard similarity, cosine similarity and euclidean similarity. Also, the loss function in the classifier is reduced by optimizing the weight parameters using the EAO approach. The performance analysis shows that the proposed recommendation model obtains improved results in terms of accuracy of 99.98%, recall of 99.81%, precision of 99.65%, and F-measure of 99.95%. The comparative analysis exhibit that the proposed EMRMR_EAO model attains better performance than the other existing works in the literature. © 2022 Elsevier B.V.

Mask Detection for Life Protection

"Health is Wealth” such a wealth of people is being affected and they are put to death on deathbeds in millions by a newly discovered virus called n-CoV (Corona virus). Covid is so-called mysterious because people affected by this disease are asymptotic. This infectious disease is mostly transmitted through droplets when an infected person coughs or sneezes. One of the main precautions that everyone must follow is wearing a mask. Some people wear masks improperly whenever they visit crowded places where there are high chances of this disease being spread. We have designed a real-time face mask detector that aims at detecting masks worn by people to reduce the transmission of this virus and make people wear masks in crowded areas. We can install this system at the entrance of places where there is more number of crowds. The detector follows Convolutional Neural Network (CNN), a part of Deep-Learning used to analyse visual imagery. It takes an input image, assigns importance (learnable weights and biases) to various aspects/objects of the image, and differentiates one from the other. Cascade Classifier detects the frontal face. Results have shown the detector classifies people with and without masks at an accuracy of 96.97%. © 2022 WMSCI.All rights reserved.

A novel COVID diagnosis and feature extraction based on discrete wavelet model and classification using X-ray and CT images.

Recently, the Covid-19 pandemic has affected several lives of people globally, and there is a need for a massive number of screening tests to diagnose the existence of coronavirus. For the medical specialist, detecting COVID-19 cases is a difficult task. There is a need for fast, cheap and accurate diagnostic tools. The chest X-ray and the computerized tomography (CT) play a significant role in the COVID-19 diagnosis. The advancement of deep learning (DL) approaches helps to introduce a COVID diagnosis system to achieve maximum detection rate with minimum time complexity. This research proposed a discrete wavelet optimized network model for COVID-19 diagnosis and feature extraction to overcome these problems. It consists of three stages pre-processing, feature extraction and classification. The raw images are filtered in the pre-processing phase to eliminate unnecessary noises and improve the image quality using the MMG hybrid filtering technique. The next phase is feature extraction, in this stage, the features are extracted, and the dimensionality of the features is diminished with the aid of a modified discrete wavelet based Mobile Net model. The third stage is the classification here, the convolutional Aquila COVID detection network model is developed to classify normal and COVID-19 positive cases from the collected images of the COVID-CT and chest X-ray dataset. Finally, the performance of the proposed model is compared with some of the existing models in terms of accuracy, specificity, sensitivity, precision, f-score, negative predictive value (NPV) and positive predictive value (PPV), respectively. The proposed model achieves the performance of 99%, 100%, 98.5%, and 99.5% for the CT dataset, and the accomplished accuracy, specificity, sensitivity, and precision values of the proposed model for the X-ray dataset are 98%, 99%, 98% and 97% respectively. In addition, the statistical and cross validation analysis is conducted to validate the effectiveness of the proposed model.

Analysis of Emerging Preprocessing Techniques Combined with Deep CNN for Lung Disease Detection

Data preprocessing is one of the pertinent steps while classifying images via CNN models. The efficiency of any model depends on the quality of the dataset it deals with. A clean dataset provides an efficient platform for a model to tackle classification and segmentation issues. Our paper focuses on three emerging data preprocessing techniques: Real ESRGAN, Swin IR, and GFPGAN over the lung disease dataset. We have used three models: Mobile net, Densenet201, and NasNet, to carry out classification tasks on Chest X-Ray images of six different types of lung disease: Bacterial Pneumonia, Viral pneumonia, Lung opacity, Covid, Tuberculosis, and Normal. Analysis of the aforementioned preprocessing techniques followed by classification via three CNN models (Mobile net, Densenet, and NasNet) are carried out on lung disease dataset, and their accuracy prediction, Training, and validation loss are extensively compared. © 2022 IEEE.

A CNN Model for Detecting Coronavirus in Chest Computed Tomography Scan Images using Gabor Filter in Pre-processing Stage

Coronavirus is the cause of the pandemic illness. The Reverse Transcription-Polymerase Chain Reaction (RT-PCR) test is frequently used to identify coronavirus. On Computed Tomography (CT) images, the extent to which the virus has impacted the lungs can be seen clearly. In 15 minutes, CT data are accessible, but RT-PCR takes 24 hours. The proposed model looks for the virus in the lungs, which is more accurate than PCR, which only looks for it in the nose or throat. More accurate and dependable data can be obtained, if Computed Tomography scans are employed. The proposed innovative model has an accuracy with Gabor filter and without Gabor filter is 0.83 and 0.75 in recognizing the coronavirus in Lung Computed Tomography Scans. The accuracy of the preceding models VGG16, VGG19, Res Net50, and Mobile Net with the Gabor filter is 0.79,0.81,0.81,0.81 and 0.68,0.61,0.71 and 0.79 without it. Gabor filter is a linear filter that is sensitive to orientation and can assist reduce noise from data. Our model obtains an accuracy of 0.83, which is higher than the Gabor Filter models VGG16, VGG19, Res Net50, and Mobile Net. © 2022 IEEE.

Mask Detection Using Convolutional Neural Network

COVID-19 virus has hit Indonesia since early March 2020. One of the government's efforts to prevent the spread of COVID-19 is to do physical distancing to require people to wear masks when doing activities outside the home. One way to overcome this problem is by detecting mask users to be more obedient and obedient to the rules, then the identification process is carried out for mask users and those who do not use masks. The process is carried out using the Convolutional Neural Network method. CNN is known to be superior and does not require pre-processing so it saves more time. In terms of algorithmic competence, CNN is considered capable of carrying out the data detection process well. Of the 1376 datasets used, 30 epochs, accuracy = 0.988, recall = 0.990, precision = 0.987, and F1 = 0.988 with the required detection time for each image between 4 to 5 seconds. © 2022 IEEE.

Pre-Processing and Enhancement Techniques for Covid-19 X-Ray and CT-Scan Medical Images

In medical diagnosis, medical imaging plays an important role, also plays a role in diagnosis and detection of chest diseases. The Pre-Processing techniques are an important step required to increase quality for Chest X-ray and CT medical images. These Techniques are used of improving the quality of the Covid19 x-ray and CT scan images. Medical images contain many unnecessary noise components in the scanned images' actual format. Some of the image preprocessing techniques are needed to eliminate certain irritating sections of an image to properly visualize the images until specifically finding the diseases.The main aim of this paper is to apply pre-processing on the images of the lungs which includes images for Covid-19, Normal and pneumonia to improve the quality of the images. Image quality improvement is accomplished by the application of filtering techniques, noise reduction and contrast enhancement techniques. The proposed technique is evaluated by using Peak signal-to-noise ratio (PSNR), Mean Square Error (MSE), and Absolute Mean Brightness Error (AMBE) to evaluate the contrast enhancement of the image that has been processed. The results show that the used technique gives better images than original images. © 2022 American Institute of Physics Inc.. All rights reserved.

An Effective Approach for Forgery Detection of Fake COVID-19 Report

With advancements in technology and image processing software, digital image forgery has become increasingly simple. However, because digital images, such as COVID-19 reports, are a common source of information, the authenticity of these digital reports has become a big concern. In recent days, it has been discovered that an increasing number of academics have begun to focus on the issue of digital report manipulation. A new deep learning-based digital picture forgery detection solution has been acquired. The mechanism is intended to ensure that the COVID-19 report on health care has not been amended or tampered with. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

E-learning course recommendation based on sentiment analysis using hybrid Elman similarity

Online learning is also referred to as E-learning which has gained huge attention and attracted most people during the COVID-19 lockdowns. Due to the excess of online information, users face severe challenges and difficulties realizing the best course that is being competitive in the global market. Therefore, it is necessary to develop an online recommendation system that supports the users in selecting the finest course with E-learning. Thus, the proposed work develops a robust RS model using different approaches. Initially, the pre-processing stage is performed to reduce the presented noise in the website data. Then, the feature extraction stage is done to extract the needed features using Improved TF-IDF, W2V (Word 2 Vector), and Hybrid N-gram. Finally, Elman Minimal Redundancy Maximum Relevance and Enhanced Aquila Optimization (EMRMR_EAO) model is proposed to provide Robust course recommendations. In this work, the ERNN method is used to classify the sentiments based on the similarity measure of the MRMR model. The top course recommendation is afforded depending on the similarity scores like Jaccard similarity, cosine similarity and euclidean similarity. Also, the loss function in the classifier is reduced by optimizing the weight parameters using the EAO approach. The performance analysis shows that the proposed recommendation model obtains improved results in terms of accuracy of 99.98%, recall of 99.81%, precision of 99.65%, and F-measure of 99.95%. The comparative analysis exhibit that the proposed EMRMR_EAO model attains better performance than the other existing works in the literature.

Abstractive Text Summarization for Covid-19 News with Data Augmentation

Since the spread of Corona Virus disease or Covid-19 at the end of 2019, there has been an extensive amount of news about Covid-19 and it takes a long time for humans to read the news, process it and retrieve important information from it. Therefore, automatic text summarization is necessary in this matter as it can help us process information faster and use it to make better decisions. Currently, there are two main approaches to automatic text summarization: extractive and ive. Extractive text summarization is conducted by identifying important parts of the text and extract a subset of sentences from the original text. ive text summarization is closer to human's method as it is the reproduction or rephrasing based on interpretation and understanding of the text using natural language processing techniques. In this paper, we present text summarization of Covid-19 news using ive method to be close to human's method of summary. We also apply data augmentation in the pre-processing part to be an example case of working with data that are not perfect or diverse enough. © 2022 IEEE.