Assessing an AI-based Smart Imagery Framing and Truthing (SIFT) system to assist radiologists annotating lung abnormalities on chest X-ray images for development of deep learning models

To assess a Smart Imagery Framing and Truthing (SIFT) system in automatically labeling and annotating chest X-ray (CXR) images with multiple diseases as an assist to radiologists on multi-disease CXRs. SIFT system was developed by integrating a convolutional neural network based-augmented MaskR-CNN and a multi-layer perceptron neural network. It is trained with images containing 307,415 ROIs representing 69 different abnormalities and 67,071 normal CXRs. SIFT automatically labels ROIs with a specific type of abnormality, annotates fine-grained boundary, gives confidence score, and recommends other possible types of abnormality. An independent set of 178 CXRs containing 272 ROIs depicting five different abnormalities including pulmonary tuberculosis, pulmonary nodule, pneumonia, COVID-19, and fibrogenesis was used to evaluate radiologists' performance based on three radiologists in a double-blinded study. The radiologist first manually annotated each ROI without SIFT. Two weeks later, the radiologist annotated the same ROIs with SIFT aid to generate final results. Evaluation of consistency, efficiency and accuracy for radiologists with and without SIFT was conducted. After using SIFT, radiologists accept 93% SIFT annotated area, and variation across annotated area reduce by 28.23%. Inter-observer variation improves by 25.27% on averaged IOU. The consensus true positive rate increases by 5.00% (p=0.16), and false positive rate decreases by 27.70% (p<0.001). The radiologist's time to annotate these cases decreases by 42.30%. Performance in labelling abnormalities statistically remains the same. Independent observer study showed that SIFT is a promising step toward improving the consistency and efficiency of annotation, which is important for improving clinical X-ray diagnostic and monitoring efficiency. © 2023 SPIE.

COVID-19 Detection from Pulmonary CT Images using Neural Networks based on Dropout-Driven Hidden Layers

Enhanced diagnosis with considerably good sensitivity and specificity is highly indispensable for COVID-19 diagnosis using radiological data to combat hazardous viral infection. Accuracy of diagnosis is a very important part that helps in further triaging and disease management. Artificial intelligent techniques using Convolutional Neural Networks and their modified alternatives have been recognized to be the salvation in chaotic situations and emergencies. Despite their immense ability to give quality results, they suffer from overfitting problems which have to be reduced by regularizing the networks. Dropout is one such regularization that modifies the network to achieve improved performance by discarding the unwanted nodes in the network layers. A simple neural network architecture inspired by former renowned architectures with dropout-driven hidden layers, CVDNN is built and experimented with for various dropout probabilities (0.1, 0.25, 0.5 and 0.75). The model was also tested with different numbers of dense layers: CVDNN1 with a single dense layer and CVDNN2 with two dense layers of a fixed dropout probability of 0.5 in it. The models are trained and tested with pulmonary computed tomography images to distinguish COVID-19 abnormality against normal cases. The CVDNN2 model presents better functioning with improved performance measures than CVDNN1 with an accuracy of 92.86 % accuracy, 90.21% sensitivity and a specificity of 95.52% for the dataset used. Dropout probabilities of 0.25 and 0.5 present reliable and better results compared to the other values experimented with. Hence a dropout-driven hidden layer can enhance the neural network's performance by choosing either 0.25 or 0.5 preferably for different applications. © 2023 IEEE.

An Efficient and Intelligent Systems for Internet of Things Based Health Observance System for Covid 19 Patients

This paper describes how an IoT -based health monitoring system was conceived and built (IoT). With the proliferation of new technologies, doctors nowadays are constantly on the lookout for cutting-edge electronic tools that will make it simpler to detect abnormalities in the human body. The Internet of Things makes it possible to create cutting-edge, non-intrusive healthcare assistance systems. In this article, we introduce the Comprehensive Health Monitoring System, or CHMS. Normal people can't afford to buy separate devices or make frequent trips to hospitals. Our CHMS will monitor a patient's vitals, including temperature, heart rate, and oxygen saturation (OS), and relay that information to a portable device. To make sense of the information gathered by the physical layer's sensors, the logical layer must analyses it. The application layer then makes judgments based on the processed data from the logical layer. The primary goal is to reduce costs for average consumers. Patients will have simple access to individual healthcare, in addition to financial sustainability. This study introduces an IoT -based system that would streamline the operation of a complex medical gadget while reducing its associated cost, allowing its users to do so from the comfort of home. The public's adoption of these gadgets as aids in a given setting might have significant effects on their own lives. © 2023 IEEE.

Enhanced Feed Forward Neural Network with Adam Optimization Model (Efnnao) for Predicting the Type 2 Diabetes Using Internet of Things

The Internet of Things (IoT) is a growing technology which connects things or objects with the internet and also enables things to collect and exchange data in the network. IoT plays a vital role in all domains, especially healthcare, where IoT used for monitoring patients and taking valuable decision for a particular problem. In the current era, diabetes is a common disease among most of the people. Diabetes is associated with many life-threatening diseases such as heart attack, kidney failure, Vison loss, Covid, etc., Type 2 diabetes is a type of diabetes that usually affects the elderly. Therefore, early detection or prediction can help prevent the patient from being at risk. However, accurately analyzing the dataset collected to make the right decision is one of the biggest tasks and improving the accuracy of the prediction model is another important task. There is several analysis models are available, over the years, various Neural Network models have been used in clinical diagnosis. However, these models are still sustained a particular level of error and less accuracy in training and testing of disease diagnosis. So, this paper proposed the Enhanced Feed forwarded Neural Network with Adam Optimization model (EFNNAO) including multiple layers of network that suitable for processing IoT based dataset. The proposed model effectively structured for predicting the type 2 diabetes in IoT environment. The designed network has the ability to learn every aspect of the dataset and perform calculations efficiently by avoiding under fitting and over-fitting. Finally, the proposed model is compared with other models which are in the same aspect. The proposed EFNNAO is outperformed than other models with 92.02% accuracy. © 2022 IEEE.

Dynamic correlations of renewable-energy companies: Evidence from a multilayer network model

Against the background of seeking to achieve carbon neutrality, relationships among renewable-energy companies around the world have become multiple and complex. In this work, the Pearson, Kendall, tail, and partial correlation coefficients were applied to 51 global companies - including solar and wind firms, independent power plants, and utilities - to explore the linear, nonlinear, extreme-risk, and direct relations between them. Sample data from 7 August 2015 to 6 August 2021 were considered, and three sub-periods were extracted from these sample data by analysis of the evolution of multiple correlations combined with event analysis. A four-layer correlation network model was then constructed. The main results are as follows. (1) The multiple relations among the selected firms underwent dramatic changes during two external shocks (the China-US trade war and the COVID-19 pandemic). (2) The extreme-risk network layer verified that the trade war mainly affected the relationships among companies in the solar industries of China and the US. (3) During the COVID-19 pandemic period, the linear and direct relationships among wind firms from Canada, Spain, and Germany were significantly increased. In this sub-period, edge-weight distributions of the four different layers were heterogeneous and varied from power-law features to Gaussian distributions. (4) During all the sub-periods, most companies had similar numbers of neighbors, while the numbers of neighbors of a few companies varied greatly in the four different layers. These findings provide a useful reference for stakeholders and may help them understand the connectedness and evolution of global renewable-energy markets. © 2023 Author(s).

Deep CNN based Multi Classification of Respiratory Disease using X-Ray Images

COVID-19 debuted in Wuhan, China on December 19, 2019. In a brief period, deadly virus now migrated to practically every country. To avoid the causative agent COVID-19 disease, governments implement a number of strict restrictions, notably prohibiting people from leaving their homes. This paper focused on detecting and classifying disease such as viral pneu-monia, covidand normal from x-ray images using deep learning methods along with pre-trained models. Moreover, validation accuracy of CNN model attained around 91 % while performing layers in neural network. Several investigations examined that identifying disease of covid reached more accuracy around 98% with hybrid and other algorithms without removing noise from particular images. But this work mainly focused on normalizing images to make the computation very efficient, convergence faster too. © 2022 IEEE.

Detecting COVID-19 on CT Images with Impulsive-Backpropagation Neural Networks

This paper presents an impulsive-backpropagation neural network (IBNN) based learning algorithm for detecting Coronavirus Disease 2019 (COVID-19), by classifying chest computed tomography (CT) images. Inspired by the nerve impulses in brain networks, the IBNN algorithm consists of two parts: a multi-layered network of impulsive neurons and a gradient decent backpropagation mechanism. The effectiveness of the IBNN algorithm is validated on clinical COVID-19 database, and a classification accuracy of 98.19% is achieved. It is further demonstrated by comparative studies that the IBNN may outperform some other learning algorithms through the integration of nerve impulses and backpropagation. Considering the intricate attributes of the chest CT scan images, the IBNN algorithm also exhibits a potential capacity of pattern recognition on complicated samples. © 2022 IEEE.

An Efficient COVID-19 Disease Outbreak Prediction Using BI-SSOA-TMLPNN and ARIMA

Globally, people's health and wealth are affected by the outbreak of the corona virus. It is a virus, which infects from common fever to severe acute respiratory syndrome. It has the potency to transmit from one person to another. It is established that this virus spread is augmenting speedily devoid of any symptoms. Therefore, the prediction of this outbreak situation with mathematical modelling is highly significant along with necessary. To produce informed decisions along with to adopt pertinent control measures, a number of outbreak prediction methodologies for COVID-19 are being utilized by officials worldwide. An effectual COVID-19 outbreaks' prediction by employing Squirrel Search Optimization Algorithm centric Tanh Multi-Layer Perceptron Neural Network (MLPNN) (SSOA-TMLPNN) along with Auto-Regressive Integrated Moving Average (ARIMA) methodologies is proposed here. Initially, from the openly accessible sources, the input time series COVID-19 data are amassed. Then, pre-processing is performed for better classification outcomes after collecting the data. Next, by utilizing Sine-centered Empirical Mode Decomposition (S-EMD) methodology, the data decomposition is executed. Subsequently, the data are input to the Brownian motion Intense (BI) - SSOA-TMLPNN classifier. In this, the diseased, recovered, and death cases in the country are classified. After that, regarding the time-series data, the corona-virus's future outbreak is predicted by employing ARIMA. Afterwards, data visualization is conducted. Lastly, to evaluate the proposed model's efficacy, its outcomes are analogized with certain prevailing methodologies. The obtained outcomes revealed that the proposed methodology surpassed the other existing methodologies. © 2023 World Scientific Publishing Company.

A Secure Wearable Framework for Stress Detection in Patients Affected by Communicable Diseases

The emergence of COVID-19 has drastically altered the lifestyle of people around the world, resulting in significant consequences on people's physical and mental well-being. Fear of COVID-19, prolonged isolation, quarantine, and the pandemic itself have contributed to a rise in hypertension among the general populace globally. Protracted exposure to stress has been linked with the onset of numerous diseases and even an increased frequency of suicides. Stress monitoring is a critical component of any strategy used to intervene in the case of stress. However, constant monitoring during activities of daily living using clinical means is not viable. During the current pandemic, isolation protocols, quarantines, and overloaded hospitals have made it physically challenging for subjects to be monitored in clinical settings. This study presents a proposal for a framework that uses unobtrusive wearable sensors, securely connected to an artificial intelligence (AI)-driven cloud-based server for early detection of hypertension and an intervention facilitation system. More precisely, the proposed framework identifies the types of wearable sensors that can be utilized ubiquitously, the enabling technologies required to achieve energy efficiency and secure communication in wearable sensors, and, finally, the proposed use of a combination of machine-learning (ML) classifiers on a cloud-based server to detect instances of sustained stress and all associated risks during times of a communicable disease epidemic like COVID-19. © 2001-2012 IEEE.

Analysis for Data Mobility and Covid-19 Positive Rate with Multilayer Perceptron

COVID-19 has been affecting human mobility to avoid the risk of infection. Movement restriction was one of the government policies to reduce the rate of infection. However, the mobility was still occurred to be recorded during the policy. This action has led to the problem of the number of beds on hospital have to be prepared for the peak of infection. This study developed a model using Multilayer perceptron as a useful theorem in regression analysis to see the fitness approximation over this problem. Five layers neural networks combination have been used to see the performance of the model to reach the best fit of the model. The process of the study includes data acquisition of the influence of community mobility over the positive number of COVID-19, managed hyperparameters, and calculate the results of prediction in the form of the length of time the patient would be infected with COVID-19 from 2020 to 2021. This study found that the infection was happening mostly after 12 days of human mobility activity in public area such as ATM, market, park, and any public area recorded by Google mobility data. It was also showed the number of infections after 12 days in order to prepare the number of beds on hospital. Furthermore, this study found the best model with smallest loss value on 0.01452617616472448 with the gap number of infection from public area as much as 77 persons. © 2022 IEEE.

Structure and Toughness Optimization of Multi-layer Weighted Regional Logistics Network: Taking the Central Region of China as an Example

The operation of the regional logistics network is often interrupted by emergencies such as rainstorms and earthquakes, especially the COVID-19 pandemic in recent years. Therefore, it is particularly important to improve the toughness of the regional logistics network to resist the risk of emergencies. This paper firstly constructed a multi-layered weighted regional logistics network of highways and railways in the central region of China based on the gravity model, analyzed its network structure characteristics by using dominant flow and social network analysis methods, then simulated the evolution trend of network toughness under different strategies. Finally, the optimization model of logistics network structural toughness under fixed cost was proposed to explore the optimization path of network structural toughness. The results show that: (1) The economically developed cities are located in the core area of the regional logistics network, on the contrary, they are located in the edge area of the regional logistics network. (2) The network as a whole has formed a "two main and four auxiliary” distribution pattern with Zhengzhou and Wuhan as the two main cores in the north and south, and Taiyuan, Hefei, Changsha, and Nanchang as the four auxiliary cores. (3) The network has higher toughness under the node random order failure strategy than under the node specified order failure strategy, and the optimization plans improve the structural toughness of the regional logistics network by 11.68%. © 2022 SPIE.

Reducing False-Positive Detections Using the Distance Between Activation Distributions in Individual Channels

We propose a novel method to estimate the confidence of outputted predictions of a convolutional neural network. We show that different channels in one layer can be treated as an ensemble and extract the confidence of a prediction from a single channel. To achieve this, we compute statistical distances between activation distributions located at the predicted mask and its surrounding area and aggregate it across all channels in a deep layer of a network. Research on a segmentation network of lung cancer nodules from 3d computer tomography images has shown growth of precision compared to the thresholding output network values. The more layers used to compute confidence, the better performance obtained, allowing for up to 18% fewer false-positives detections on the source Cancer dataset and up to 54% fewer false-positives detections on an unseen Covid dataset. Analyzing channel activations doesn't require any changes in the training procedure with a negligible amount of additional computations at the inference time. © 2022 IEEE.

Multiple Transfer Learning-AI design for Lung Disease-Pneumothorax and Tension Pneumothorax classification from Chest-X ray

The research paper discuss the Artificial Intelligence based Multiple Transfer Learning Mechanism in identification of lung diseases like pneumothorax, tension pneumothorax from a set of chest X-rays. Pneumothorax being a primary stage of many sorts of pulmonary diseases, it has now a days being noticed as an impact with COVID cases due to the insertion of the tubes into the lungs. The proper diagnosis of the various stages of Pneumothorax is thus essential in the current scenario. Identification of the patients with Pneumothrax with less diagnostic time is the highlight of this research work. The deep learning technology of AI has enlightened the research in the medical imaging field. The chest X-ray images are with the pre-processing analysis, normalised the images for a uniform image data processing. The advanced method of transfer learning is equipped with modifications in the various fully connected convolutional network layers. The modified transfer learning has been used with DenseNet and VGG 19. The convolutional neural networks with DenseNet201 and VGG19 utilized stochastic gradient decent optimization for parameter optimization. The data set with pneumothorax and tension pneumothorax along with the control set has been trained and validated. The training and validation of these network has proven results with 89% accuracy with VGG19 and 100% accuracy with Densenet. The evaluation of modified Multi-transfer learning algorithm is identified successfully with new random input chest X-ray with a less diagnostic time. © 2022 IEEE.

Identification of Coronavirus & Pneumonia from Chest x-ray Scans using Deep Multilayered CNN

Recently, COVID-19 disease carried out by the SARS-CoV-2 virus appeared as a pandemic across the world. The traditional diagnostic techniques are facing a hard time detecting the virus efficiently at an early stage. In this context, chest x-ray scans can be useful for diagnostic prediction. Therefore, in this paper, a deep multi-layered convolution neural network has been proposed to analyze the chest x-ray scans effectively for detecting COVID-19 and pneumonia accurately. The proposed approach has been applied on multiple benchmark datasets and the experimental results define the effectiveness of the proposed approach. © 2021 IEEE.

Dynamic Analysis of Demographic Sentiment

There is no doubt that big data analysis has a very positive impact on economics, security, and other aspects for countries and enterprises alike. Where we have recently noticed the frantic competition between companies to increase their profits by analyzing the largest amount of data as quickly as possible. Especially analyzing data related to Covid-19 to make the most of information in all areas. Covid-19 has drastically affected many lives in recent years but, even in these hard times, businesses can leverage the current pandemic to make a profit. In this paper, we investigate a variety of tweets using MapReduce, Spark, and Machine Learning methods to determine the sentiment of a given tweet based on the information provided by the dataset. With this information, businesses could learn how to present Covid-19 and pandemic related goods and information in a way that will be well received by its audience. To take this a step further, we will investigate trends in sentiment across demographics tweeting about the virus. This information in sentiment is dynamically useful to understand how specific audiences feel about the pandemic. We explore which Machine Learning methods produce the best results such as Multi-Layer Perceptron neural networks and Logistic Regression. © 2022 IEEE.

Detection of COVID-19 and Prediction of Pneumonia from Chest X-Rays using Deep Learning

The occurrence of COVID-19 disease and the pandemic has caused severe detrimental effects on global health, economy, and well-being. Ignorance and disobeying precautionary measures result in the rapid spread of this contagious infection and subsequent illnesses. There have been more than forty crore cases and fifty lakh deaths reported worldwide. Though the RT-PCR manual testing is widely carried out for diagnosing the presence of the virus in the human body, it is less reliable than chest CT and X-Ray imaging techniques. These tests are swift and assist in determining the severity of the infection with affordable costs. Hence, recent advances involve the application of artificial intelligence and deep learning techniques for automatic detection of this disease. Various datasets of X-Ray images of the chest are available online which are utilized for training and validation of results. In this paper, a dense layer Convolutional Neural Network with one twenty one layers model is used to predict the pneumonia type from the Chest X-Ray images. The results acquired from the implementation show that the model has the highest accuracy of 0.9738 while the specificity is 93.6%, and it outperforms the other similar models. Hence, the possibility of a false positive case occurring is lower and can facilitate easy diagnosis of the infection. © 2022 IEEE.

Experimental Investigation and neural network modelling of diesel engine using hexanol blended ternary waste cooking oil biodiesel with moderate preheating

The depleting fossil fuel reserves, rising air pollution, technology transformation threat, and most recently, global economic slowdown by the COVID-19 pandemic, led the internal combustion engine-based automotive industries in a critical condition. The development of improved biofuels to meet stringent emission norms is a promising solution. Higher alcohols possess the fuel properties better than lower alcohols to blend with diesel and biodiesel. The miscibility and higher viscosity is the issue. Preheating can help the vaporization and atomization of fuel. The present study investigates the engine characteristics of moderately preheated ternary fuel using 20 to 40% blends of 1-hexanol, waste cooking oil biodiesel, and diesel. The study found that moderately preheated ternary fuel blends showed a drop in brake-specific fuel consumption, HC, CO, and smoke emissions with improvement in peak cylinder pressure, heat release rate, and brake thermal efficiency. A multi-layer neural network model is developed to prognosticate the engine characteristics. Backpropagation algorithm-based neural network with single hidden layers using Levenberg–Marquardt training function gave the best results. The mean square error of the network was 0.00028517 and the correlation coefficient was 0.99944, 0.99945, and 0.99923 for training, validation, and testing respectively. The mean absolute percentage error was found below 4%. © 2022 Elsevier Ltd

Multi Layered Deep Neural Network for Feature Extraction in Cross Domain Crowd Counting

Automated crowd density monitoring is an emerging area of research. It is a vital technology that assists during recent disease outbreaks in preserving social distancing, crowd management and other widespread applications in public security and traffic control. Modern methods to count people in crowded scenes mainly rely on Convolutional Neural Network (CNN) based models. But the model's ability to adapt for different domains which is referred to as cross domain crowd counting is a challenging task. To remedy this difficulty, many researchers used Spatial Fully Convolutional Network (SFCN) based crowd counting models with synthetic crowd scene data. They covered many image domains with few-shot learning to reduce the domain adaptation gap between source and target image domains. In this paper, we propose a new multi-layered model architecture instead of SFCN single-layered model architecture. The proposed model extracts more meaningful features in image scenes along with large scale variations to increase the accuracy in cross domain crowd counting. Furthermore, with extensive experiments using four real-world datasets and analysis, we show that the proposed multi-layered architecture performs well with synthetic image data and few-shot learning in reducing domain shifts. © 2022 IEEE.

The Experience with the REU-sponsored Project on Predicting COVID-19 Pandemics Using Physics-Guided Graph Attention Networks

The COVID-19 pandemic has significantly impacted most countries in the world. Analyzing COVID-19 data from these countries together is a prominent challenge. Under the sponsorship of NSF REU, this paper describes our experience with a ten-week project that aims to guide a REU scholar to develop a physics-guided graph attention network to predict the global COVID-19 Pandemics. We mainly presented the preparation, implementation, and dissemination of the addressed project. The COVID-19 situation in a country could be dramatically different from that of others, which suggests that COVID-19 pandemic data are generated based on different mechanisms, making COVID-19 data in different countries follow different probability distributions. Learning more than one hundred underlying probability distributions for countries in the world from large scale COVID-19 data is beyond a single machine learning model. To address this challenge, we proposed two team-learning frameworks for predicting the COVID-19 pandemic trends: peer learning and layered ensemble learning framework. This addressed framework assigns an adaptive physics-guided graph attention network (GAT) to each learning agent. All the learning agents are fabricated in a hierarchical architecture, which enables agents to collaborate with each other in peer-to-peer and cross-layer way. This layered architecture shares the burden of large-scale data processing on machine learning models of all units. Experiments are run to verify the effectiveness of our approaches. The results indicate the proposed ensemble outperforms baseline methods. Besides documented on GitHub, this work has resulted in two journal papers. © 2022 IMCIC 2022 - 13th International Multi-Conference on Complexity, Informatics and Cybernetics, Proceedings. All rights reserved.

Smart Infrared Temperature Measurement System Based On Elastic Compute Service Platform

In the fight against the novel coronavirus, this paper designs a smart infrared temperature measurement system based on the Elastic Compute Service platform. In the perceptual layer part of the Internet of Things(IoT), the system uses infrared temperature sensors to quickly collect the temperature of the forehead or arm of the human body and Uses the MCU to automatically transmits the measured data to the Elastic Compute Service through the WiFi module. In the network layer part of the Internet of Things,the data written into the database of the Elastic Compute Service through the program deployed on the Elastic Compute Service. In the application layer of the Internet of Things, the remote management terminal monitors the collected body temperature data in real-time and provides quick warnings. Finally, the User can log in to the applet on his mobile phone to quickly and easily obtain personal information. © 2021 IEEE