Performance evaluation of regression models for COVID-19: A statistical and predictive perspective

Prospective virtual screening combined with bio-molecular simulation enabled identification of new inhibitors for the KRAS drug target

Lung cancer is a disease with a high mortality rate and it is the number one cause of cancer death globally. Approximately 12-14% of non-small cell lung cancers are caused by mutations in KRASG12C. The KRASG12C is one of the most prevalent mutants in lung cancer patients. KRAS was first considered undruggable. The sotorasib and adagrasib are the recently approved drugs that selectively target KRASG12C, and offer new treatment approaches to enhance patient outcomes however drug resistance frequently arises. Drug development is a challenging, expensive, and time-consuming process. Recently, machine-learning-based virtual screening are used for the development of new drugs. In this study, we performed machine-learning-based virtual screening followed by molecular docking, all atoms molecular dynamics simulation, and binding energy calculations for the identifications of new inhibitors against the KRASG12C mutant. In this study, four machine learning models including, random forest, k-nearest neighbors, Gaussian naïve Bayes, and support vector machine were used. By using an external dataset and 5-fold cross-validation, the developed models were validated. Among all the models the performance of the random forest (RF) model was best on the train/test dataset and external dataset. The random forest model was further used for the virtual screening of the ZINC15 database, in-house database, Pakistani phytochemicals, and South African Natural Products database. A total of 100 ns MD simulation was performed for the four best docking score complexes as well as the standard compound in complex with KRASG12C. Furthermore, the top four hits revealed greater stability and greater binding affinities for KRASG12C compared to the standard drug. These new hits have the potential to inhibit KRASG12C and may help to prevent KRAS-associated lung cancer. All the datasets used in this study can be freely available at ( https://github.com/Amar-Ajmal/Datasets-for-KRAS ).

Design and hierarchical analysis of magnetic actuated robot: A governing equation based approach

As the alternative solution to the conventional guidewire, the magnetic robot can help interventionists perform percutaneous coronary intervention (PCI) because magnetic fields are transparent and safe for biological tissues. Despite extensive research on magnetic robots, the exploration of their deflection control for practical applications still requires further research. In this paper, a hierarchical analysis framework (HAF) is proposed to control the magnetic robot's deflection. Six deflection subpatterns are analyzed through HAF, incorporating statistical and regression analyses to establish governing equations of magnetic robots. The performance of the control equations is validated through precise control of the magnetic continuum robot (MCR) and magnet-tipped robot (MTR) in both uniform and gradient magnetic fields. Experimental results show that under the uniform magnetic field, the average root mean square error (RMSE) of governing equation of MCR is 0.08±0.05°, 0.41±0.34°, 1.47±0.49° and 1.07±0.66° for four-types horizontal deflection, 0.19±0.07mm and 0.16±0.10mm for two-types vertical deflection, respectively. Based on the governing equations, the MTR is able to precisely navigate to coronary arteries with various degrees of stenosis (30%, 52%, and 60%), and successfully pass through a series of rings, with an average error of 1.05 mm. The research successfully demonstrates the potential of HAF in creating robust and reliable governing equations for magnetic actuation in medical robotics, with significant implications for enhancing the precision and safety of PCI procedures.

An Optimized Model Based on Deep Learning and Gated Recurrent Unit for COVID-19 Death Prediction

The COVID-19 epidemic poses a worldwide threat that transcends provincial, philosophical, spiritual, radical, social, and educational borders. By using a connected network, a healthcare system with the Internet of Things (IoT) functionality can effectively monitor COVID-19 cases. IoT helps a COVID-19 patient recognize symptoms and receive better therapy more quickly. A critical component in measuring, evaluating, and diagnosing the risk of infection is artificial intelligence (AI). It can be used to anticipate cases and forecast the alternate incidences number, retrieved instances, and injuries. In the context of COVID-19, IoT technologies are employed in specific patient monitoring and diagnosing processes to reduce COVID-19 exposure to others. This work uses an Indian dataset to create an enhanced convolutional neural network with a gated recurrent unit (CNN-GRU) model for COVID-19 death prediction via IoT. The data were also subjected to data normalization and data imputation. The 4692 cases and eight characteristics in the dataset were utilized in this research. The performance of the CNN-GRU model for COVID-19 death prediction was assessed using five evaluation metrics, including median absolute error (MedAE), mean absolute error (MAE), root mean squared error (RMSE), mean square error (MSE), and coefficient of determination (R2). ANOVA and Wilcoxon signed-rank tests were used to determine the statistical significance of the presented model. The experimental findings showed that the CNN-GRU model outperformed other models regarding COVID-19 death prediction.

A Regression Analysis on Steam Gasification of Polyvinyl Chloride Waste for an Efficient and Environmentally Sustainable Process

Over the last few years, researchers have shown a growing interest in polyvinyl chloride (PVC) gasification and have conducted several studies to evaluate and enhance the process. These studies have recognized that processing parameters have a crucial impact on the assessment of PVC gasification. Despite this, there has been limited exploration of the use of machine learning techniques, particularly regression models, to optimize PVC waste gasification. This study aims to investigate the effectiveness of regression models as machine learning algorithms in predicting the performance of PVC waste gasification. The study uses data collected through a validated thermodynamic model, and three different regression models are tested and compared in detail. Cold gas efficiency and normalized carbon dioxide emission are predicted using linear, quadratic, and quadratic with interaction algorithms. The outcomes for emission algorithms reveal that the linear emission algorithm possesses a high R-square value of 97.49%, which indicates its strong predictive capability. Nevertheless, the quadratic algorithm outperforms it, exhibiting an R-square value of 99.81%. The quadratic algorithm with an interaction term, however, proves to be the best among them all, displaying a perfect R-square value of 99.90%. A similar observation is detected for the cold gas efficiency algorithms. These findings suggest that the quadratic algorithm with an interaction term is superior and has a greater predictive accuracy. This research is expected to provide valuable insight into how regression algorithms can be used to maximize the efficiency of PVC waste gasification and reduce its associated environmental concerns.

Effects of COVID-19 pandemic on spatial preferences and usage habits of users in shopping malls and its relation with circulation layout

The COVID-19 pandemic, with its global impact on every scale of life, has caused a change in all the habits of people regarding daily life, and this change has also been directly reflected in the human-space interaction. The bidirectional relationship between human and space has evolved by being influenced by each other during the pandemic process. Within the scope of this study, which aims to determine the predicted change in the spatial preferences and usage habits of the users in the shopping malls due to the COVID-19 pandemic, which causes changes in the space and the usage habits of the users from micro scale to macro scale, analyzes were made with the research methodology consisting of three stages (i. space syntax, ii. survey, iii. comparison) and the results were evaluated. According to the results obtained from the case study, it was determined that the COVID-19 pandemic deeply changed people's spatial perception, preferences and usage habits in shopping malls and circulation layout also plays a role on it. To summarize, while personal preferences were at the forefront in places to be preferred for shopping before the pandemic, the characteristics of the place gained importance after the pandemic.

The effects of COVID-19 on safety practices in construction projects

The coronavirus (COVID-19) pandemic was recognized as a worldwide epidemic and classed as a high risk in early 2020, affecting people's health, economies, and business sectors considerably. This pandemic has had an impact on people's lifestyles and work processes in a multitude of sectors. The construction industry is one such industry that has had a substantial influence on it. However, this influence needs to be measured in different areas. This study aims to measure the effects of the COVID-19 on the 7 core safety elements and their 24 procedures that are derived from the recommended practices for safety and health programs in construction that is issued by OSHA to see whether their priorities have been changed or not. The data were collected and then analyzed using Relative Importance Index (RII) to study the changes in their priorities; and using t-test to study the significance of the changes before and after COVID-19 pandemic.

Analyzing the research trends of COVID-19 using topic modeling approach

Purpose

The COVID-19 pandemic has impacted 222 countries across the globe, with millions of people losing their lives. The threat from the virus may be assessed from the fact that most countries across the world have been forced to order partial or complete shutdown of their economies for a period of time to contain the spread of the virus. The fallout of this action manifested in loss of livelihood, migration of the labor force and severe impact on mental health due to the long duration of confinement to homes or residences.

Design/methodology/approach

The current study identifies the focus areas of the research conducted on the COVID-19 pandemic. Abstracts of papers on the subject were collated from the SCOPUS database for the period December 2019 to June 2020. The collected sample data (after preprocessing) was analyzed using Topic Modeling with Latent Dirichlet Allocation.

Findings

Based on the research papers published within the mentioned timeframe, the study identifies the 10 most prominent topics that formed the area of interest for the COVID-19 pandemic research.

Originality/value

While similar studies exist, no other work has used topic modeling to comprehensively analyze the COVID-19 literature by considering diverse fields and domains.

COVID-19 Spatio-Temporal Evolution Using Deep Learning at a European Level

COVID-19 evolution imposes significant challenges for the European healthcare system. The heterogeneous spread of the pandemic within EU regions elicited a wide range of policies, such as school closure, transport restrictions, etc. However, the implementation of these interventions is not accompanied by the implementation of quantitative methods, which would indicate their effectiveness. As a result, the efficacy of such policies on reducing the spread of the virus varies significantly. This paper investigates the effectiveness of using deep learning paradigms to accurately model the spread of COVID-19. The deep learning approaches proposed in this paper are able to effectively map the temporal evolution of a COVID-19 outbreak, while simultaneously taking into account policy interventions directly into the modelling process. Thus, our approach facilitates data-driven decision making by utilizing previous knowledge to train models that predict not only the spread of COVID-19, but also the effect of specific policy measures on minimizing this spread. Global models at the EU level are proposed, which can be successfully applied at the national level. These models use various inputs in order to successfully model the spatio-temporal variability of the phenomenon and obtain generalization abilities. The proposed models are compared against the traditional epidemiological and Autoregressive Integrated Moving Average (ARIMA) models.

Machine Learning, Deep Learning, and Mathematical Models to Analyze Forecasting and Epidemiology of COVID-19: A Systematic Literature Review

COVID-19 is a disease caused by SARS-CoV-2 and has been declared a worldwide pandemic by the World Health Organization due to its rapid spread. Since the first case was identified in Wuhan, China, the battle against this deadly disease started and has disrupted almost every field of life. Medical staff and laboratories are leading from the front, but researchers from various fields and governmental agencies have also proposed healthy ideas to protect each other. In this article, a Systematic Literature Review (SLR) is presented to highlight the latest developments in analyzing the COVID-19 data using machine learning and deep learning algorithms. The number of studies related to Machine Learning (ML), Deep Learning (DL), and mathematical models discussed in this research has shown a significant impact on forecasting and the spread of COVID-19. The results and discussion presented in this study are based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Out of 218 articles selected at the first stage, 57 met the criteria and were included in the review process. The findings are therefore associated with those 57 studies, which recorded that CNN (DL) and SVM (ML) are the most used algorithms for forecasting, classification, and automatic detection. The importance of the compartmental models discussed is that the models are useful for measuring the epidemiological features of COVID-19. Current findings suggest that it will take around 1.7 to 140 days for the epidemic to double in size based on the selected studies. The 12 estimates for the basic reproduction range from 0 to 7.1. The main purpose of this research is to illustrate the use of ML, DL, and mathematical models that can be helpful for the researchers to generate valuable solutions for higher authorities and the healthcare industry to reduce the impact of this epidemic.