ABSTRACT
This study proposes a graph convolutional neural networks (GCN) architecture for fusion of radiological imaging and non-imaging tabular electronic health records (EHR) for the purpose of clinical event prediction. We focused on a cohort of hospitalized patients with positive RT-PCR test for COVID-19 and developed GCN based models to predict three dependent clinical events (discharge from hospital, admission into ICU, and mortality) using demographics, billing codes for procedures and diagnoses and chest X-rays. We hypothesized that the two-fold learning opportunity provided by the GCN is ideal for fusion of imaging information and tabular data as node and edge features, respectively. Our experiments indicate the validity of our hypothesis where GCN based predictive models outperform single modality and traditional fusion models. We compared the proposed models against two variations of imaging-based models, including DenseNet-121 architecture with learnable classification layers and Random Forest classifiers using disease severity score estimated by pre-trained convolutional neural network. GCN based model outperforms both imaging-only methods. We also validated our models on an external dataset where GCN showed valuable generalization capabilities. We noticed that edge-formation function can be adapted even after training the GCN model without limiting application scope of the model. Our models take advantage of this fact for generalization to external data. © 2023 IEEE.
ABSTRACT
Purpose: The new challenge in Artificial Intelligence (AI) is to understand the limitations of models to reduce potential harm. Particularly, unknown disparities based on demographic factors could encrypt currently existing inequalities worsening patient care for some groups. Method(s): Following PRISMA guidelines, we present a systematic review of 'fair' deep learning modeling techniques for natural and medical image applications which were published between year 2011 to 2021. Our search used Covidence review management software and incorporates articles from PubMed, IEEE, and ACM search engines and three reviewers independently review the manuscripts. Result(s): Inter-rater agreement was 0.89 and conflicts were resolved by obtaining consensus between three reviewers. Our search initially retrieved 692 studies but after careful screening, our review included 22 manuscripts that carried four prevailing themes;'fair' training dataset generation (4/22), representation learning (10/22), model disparity across institutions (5/22) and model fairness with respect to patient demographics (3/22). We benchmark the current literature regarding fairness in AI-based image analysis and highlighted the existing challenges. We observe that often discussion regarding fairness are limited to analyzing existing bias without further establishing methodologies to overcome model disparities. Conclusion(s): Based on the current research trends, exploration of adversarial learning for demographic/camera/institution agnostic models is an important direction to minimize disparity gaps for imaging. Privacy preserving approaches also present encouraging performance for both natural and medical image domain. Copyright © 2022, Taiwanese Society of Biomedical Engineering.
ABSTRACT
With the rise of Covid-19, the importance of health monitoring has risen to a new peak. Keeping a check on the symptoms of covid is an integral part of our lifestyle now. Using Tele-Health systems can quickly achieve this feat. The Tele-Health field has vastly improved in the span of the uprise of the pandemic and has helped provide medical and non-medical individuals with the help they require. Much work has been done in this field, integrating IoT with the medical field to monitor an individual’s physical parameters efficiently and safely remotely. We have done a systematic review of the works that have helped develop this field during the pandemic. Bringing forward the pros and cons of these systems, we try to draw a clear picture to clearly understand the systems that have helped improve our daily lifestyle over this pandemic period. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
ABSTRACT
The coronavirus pandemic which spread from Wuhan China toward the latter part of 2019 has resulted in 216,303,376 confirmed cases and 4,498,451deaths deaths to date. The novelty and lack of a definitive treatment protocol for the virus and the acute respiratory distress syndrome it produces has resulted in patients being placed on artificial ventilation and most often never recovering therefrom. Very little is known about the pathophysiology of the virus and the biological mechanisms in which it disrupts to bring about the now identified wide array of clinical features which are not solely isolated to the respiratory tract. It is now an established fact however, that one of the major pathways implicated and on which often results in the death and or severe complications in COVID-19 patients is the cytokine storm. The use of new drugs to combat such a cytokine storm is thus important considering the current global COVID-19 situation so as to stop the further progression of the disease in patients and decrease both morbidity and mortality by crippling a major mechanism which hastens death in the hosts. It is, therefore, vital that a systematic analysis and review of the various therapeutic agents are undertaken to select the best drug for the treatment of patients with cytokine storm. This research aims to relate the best therapeutic regimens currently available precisely and concisely to physicians so as to ensure the best possible treatment modality is selected for each patient. An extensive review of the literature was done on the following databases: Google scholar, Trip database, EMBASE, PubMed, and PubMed Central. The keywords and the Boolean operators used for searches were "COVID-19" OR "SARS-CoV-2" AND "Therapeutics" OR "drug therapy" AND "Cytokine Release Syndrome." The discovery and the use of such drugs, namely, Tocilizumab and potent corticosteroids such as dexamethasone and methylprednisolone in the maximum daily doses of 6 mg and 250 mg, respectively, have shown positive outcome to combat cytokine storm in severe COVID-19 patients. The rationale behind the use of these drugs being to suppress the immune system and thus decrease the detrimental cytokine cascade induced in severely ill COVID-19 patients will be instrumental in the treatment and prevention of severe complication. It is vital for the various drugs under trial and implemented in emergency use to be compared and studied so as to best select the drug which can be incorporated into a treatment regimen which is both effective and has diminished adverse effects.
ABSTRACT
We investigate the binding interactions of synthesized multi-walled carbon nanotubes (MWCNTs) with SARS-CoV-2 virus. Two essential components of the SARS-CoV-2 structure i.e.6LU7 (main protease of SARS-CoV-2) and 6LZG (spike receptor-binding domain complexed with its receptor ACE2) were used for computational studies. MWCNTs of different morphologies (zigzag, armchair and chiral) were synthesized through a thermal chemical vapour deposition process as a function of pyrolysis temperature. A direct correlation between radius to volume ratio of the synthesized MWCNTs and the binding energies for all three (zigzag, armchair and chiral) conformations were observed in our computational studies. Our result suggests that MWCNTs interact with the active sites of the main protease along with the host angiotensin-converting enzyme2 (ACE2) receptors. Furthermore, it is also observed that MWCNTs have significant binding affinities towards SARS-CoV-2. However, the highest free binding energy of −87.09 kcal mol−1 with 6LZG were shown by the armchair MWCNTs with SARS-CoV-2 through the simulated molecular dynamic trajectories, which could alter the SARS-CoV-2 structure with higher accuracy. The radial distribution function also confirms the density variation as a function of distance from a reference particle of MWCNTs for the study of interparticle interactions of the MWCNT and SARS-CoV-2. Due to these interesting attributes, such MWCNTs could find potential application in personal protective equipment (PPE) and diagnostic kits. Investigation of the binding interactions of synthesized multi-walled carbon nanotubes (MWCNTs) with SARS-CoV-2 virus.
ABSTRACT
With the advancement of the Internet of Things in our smart environment, smart devices are working without human intervention. So home can be converted to intelligent home automation systems to perform its computation automatically. In a pandemic situation, the majority of people have spent their maximum time at home. So indoor air quality, insider's and outsider's health monitoring has become an important issue. As respiratory diseases are the main concern for pandemics, we have to develop an intelligent home system model to monitor healthy environmental conditions for the users. This paper proposes an energy-efficient smart system model to monitor the health and environmental condition by measuring the carbon monoxide threat level that indirectly affects other atmospheric parameters. Our system alerts when the carbon monoxide level exceeds the safe level. Remote monitoring of the home and health parameters is done in real-time with the help of the system model. For this purpose, we are adopting Dempster-Shafer evidence theory as a mathematical model to aggregate the data coming from different sensors. The sensor nodes track the home and health parameters such as room temperature, humidity, carbon monoxide level, SpO2 level, body temperature, and pulse rate. The smartphone app updates the user's real-time sensor data through the display and indirectly helps to maintain the physical distance. The proposed intelligent home-health system model is compact, cost-effective, energy-efficient for the user, and is especially useful for the quarantined covid affected people in a pandemic situation. © 2022 ACM.
ABSTRACT
The article has intended to study the action of Twitter-based media advocacy promoted by the Ministry of Health (MOH) of the Government of India, and World Health Organization (WHO) during the Covid-19 pandemic. Its goal was to assess the degree of the WHO and MOH's media campaigning for Covid-19, as well as the public's perception of this advocacy. In this regard, mixed methods have been used for data collection where a survey has been conducted with 125 respondents, who use Twitter, from Kolkata (India) with the help of random sampling. A content analysis of two well-known Twitter accounts was conducted, which helped to reflect the current trends that they follow. The findings of this research have reflected the choice of medium preferred by the respondents for receiving news and information during the Covid-19 pandemic. It has also helped to identify the Twitter handles and tweets they mostly follow and thereby the major factors influencing their choice. The outcome of this research has helped to study whether Twitter can be used for institutionalized health communication or not in the future. © 2021 The Author/s.
ABSTRACT
The spread of the Covid-19 has presented an unparalleled challenge for media management as well as for the media content. The pattern of daily life changed due to the excessive use of media. India, as a nation has been in the third position worldwide, many deaths during a pandemic are concerned. Kolkata being one of the metro cities of the country has not been exempted. The regional media content perceived a knowledge gap with the highest circulated national daily of the country. The changed media content, and audience perception towards the change, and the need for media advocacy during any health crisis in general and Covid-19 in particular, is studied in this paper using a mixed approach of both quantitative and qualitative. The discourse analysis of the newspapers in a constructed week format, representing a six months study during the pandemic, and the primary data from the audience suggested the behavior change and attitude formation through media, in this unique study. © 2021 The Author/s.
ABSTRACT
A computational model for nucleation and growth of iron (II) oxide nanoparticle (IONP) in thermal plasma has been developed. A nondimensional form of the aerosol general dynamic equations (GDEs) along with a discrete volume sectional model assumption is used to numerically solve the coupled system of GDEs. The variation in supersaturation ratio and the mean particle diameter of IONPs with respect to temperature across the plasma reactor has been presented. The scatter plot showing the distribution of particle number density of certain size across the reactor chamber is shown. In silico molecular docking study was performed to reveal the putative interaction of the IONPs with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) virus. The results revealed significant binding affinity of IONPs with 6LZG (spike receptor-binding domain complexed with its receptor ACE2) and 5RH4 (main protease) of SARS-COV-2 by forming hydrogen and hydrophobic bonds with nearby amino acid residues. The interactions of IONPs are associated with the conformational changes in the protein which could be used to treat and control SARS-CoV-2 infection.
ABSTRACT
Our work investigates the interaction of synthesized graphene with the SARS-CoV-2 virus using molecular docking and molecular dynamics (MD) simulation method. The layer dependent inhibitory effect of graphene nanosheets on spike receptor-binding domain of 6LZG, complexed with host receptor i.e. angiotensin-converting enzyme 2 (ACE2) of SARS-CoV-2 was investigated through computational study. Graphene sample was synthesized using mechanical exfoliation with shear stress and its mechanism of inhibition towards the SARS-CoV-2 virus was explored by molecular docking and molecular dynamics (MD) simulation method. The thermodynamics study for the free binding energy of graphene towards the SARS-CoV-2 virus was analyzed. The binding energy of graphene towards the virus increased with an increasing number of layers. It shows the highest affinity of -17.5 Kcal/mol in molecular docking while ΔGbinding is in the order of -28.01 ± 0.04 5 Kcal/mol for the seven-layers structure. The increase in carbon layers is associated with an increasing number of edge sp3 -type carbon, providing greater curvature, further increase the surface reactivity responsible for high binding efficiency. The MD simulation data reveals the high inhibition efficiency of the synthesized graphene towards SARS-CoV-2 virus which would help to design future in-vitro studies. The graphene system could find potential applications in personal protective equipment and diagnostic kits.Communicated by Ramaswamy H. Sarma.