Development of a Multi-modal Severity Prediction System for Covid-19 using Machine Learning Algorithms

Medical systems all over the world have been devastated by the covid19 pandemic. Even abundant and wealthy countries have struggled a lot. As of August, 2022, number of corona virus cases has been reached to almost 588 million worldwide reported to WHO. With automation at the level of covid19 severity prediction can improve healthcare delivery in parts of the world where access to skilled experts is limited. It can also help in resource management and reducing mortality rate. Method: In this research, the researchers designed and developed a novel multimodal framework for covid19 severity prediction with a high precision capacity including decisions from medical imaging and clinical factors including patient details, co morbidities and blood results. The researchers explored oversampling methods SMOTE and ROC with SVM, Decision Tree, Random Forest and ANN classifiers for predicting severity using clinical factors. Image enhancement methods gamma correction and AHE explored with ChexNet model for severity prediction through X-ray images. Performance of the predictions has been evaluated using accuracy, precision, sensitivity, and F1-score. Results: The researchers achieved superior prediction using RF classifier with SMOTE method for text dataset with accuracy of 96%. For X-ray image dataset ChexNet with AHE achieved 87% accuracy. Infection severity inversely proportional to clinical factors LYP, LY,MOP,CA, ALB and ALG where as it is directly proportional to AST, ALT,DD,CRP,LDH,BUN,CR,MCH,GLU,TBIL and WBC. In the future, performance of the image model may be improved by concatenating multi scale features from different layers of CNN to increase representation power of the CNN model. Again channel attention may be beneficial to improve model performance. © 2022, Ismail Saritas. All rights reserved.

A Framework of Smart Dispensing Unit for Optimizing Sanitizer Quantity using IoT schematics

Nowadays, technology is helping everyone to cover their basic hygiene needs. During the COVID'19 pandemic, every person has started taking precautions such as hand sanitization, wearing of masks and social distancing. In this paper, the authors are focusing on hand sanitization. Latest sanitizer machines dispense sanitizer automatically because of touchless technology, but excessive use of sanitizer is harmful to the human skin. To address this issue, a smart sanitizer dispenser based on the Internet of Things (IoT) and Machine Learning (ML) has been proposed to fight against COVID'19. The proposed solution captures the image of hands and based on the size of hands and the number of germs present on the hands, the machine will dispense the sanitizer. In this way, the damage to human skin caused due to the excessive use of sanitizer can be reduced and the amount of sanitizer can also be saved. © The Electrochemical Society

Possible opportunities in face mask manufacturing during the COVID-19 Pandemic: A study from India

COVID-19 has been primarily regarded as a respiratory disease, and until a safer and effective treatment or vaccine becomes available, the prevention of COVID-19 may continue through interventions based on non-pharmaceutical measures such as maintaining of physical distances and use of personal protective equipment like facemasks, etc. Therefore, an attempt was made in this study to explore the drawbacks with the presently available facemasks for protection from COVID-19 viruses in the state of Odisha in India, and also to explore the possible opportunities for further development of these facemasks. The associated discomforts;strength, weaknesses, opportunities, and threats (SWOT) analysis of existing facemasks in Odisha;possible opportunities for "Make in India"of these facemasks;along with safer use have been analyzed with the help of interpretive structural modelling (ISM) approach followed by MICMAC analysis. Copyright © 2022, IGI Global.

Use of the Wearable Device in Telehealth during Covid-19

The purpose of the review is to find out the extend of wearable devices in the field of telehealth and other technologies used to implement telehealth, as this was the immediate solution to the COVID-19 pandemic for maintaining the social distance and also to treat the patients for disease or consult them, without the physical presence. In the research systematic literature review is being followed. Various databases like pubmed, medline and Google scholar is used to extract information for the list of papers downloaded. Screening based on filters was done removing papers other than English language and included only research papers, excluding articles editorial, etc. Also, some papers were added by snowballing technique while going further with the research. The review helps find some great telehealth technological findings, like wearable technologies and data accumulation. The research has been followed by 28 papers and focuses on the enablers of telehealth and the implication of wearable technologies in times of COVID-19. Originality: The independent search on telehealth and wearable technologies has been combined with reviewing this study and building the framework and the existing studies have been used for the review.

Development of a classifier with analysis of feature selection methods for COVID-19 diagnosis

Purpose The COVID-19 pandemic situation is increasing day by day and has affected the lifestyle and economy worldwide. Due to the absence of specific treatment, the only way to control a pandemic is by stopping its spread. Early identification of affected persons is urgently in demand. Diagnostic methods applied in hospitals are time-consuming, which delay the identification of positive patients. This study aims to develop machine learning-based diagnosis model which can predict positive cases and helps in decision-making. Design/methodology/approach In this research, the authors have developed a diagnosis model to check coronavirus positivity based on an artificial neural network. The authors have trained the model with clinically assessed symptoms, patient-reported symptoms, other medical histories and exposure data of the person. The authors have explored filter-based feature selection methods such as Chi2, ANOVA F-score and Mutual Information for improving performance of a classification model. Metrics used to evaluate performance of the model are accuracy, precision, sensitivity and F1-score. Findings The authors got highest classification performance with model trained with features ranked according to ANOVA FS method. Highest scores for accuracy, sensitivity, precision and F1-score of predictions are 0.93, 0.99, 0.94 and 0.93, respectively. The study reveals that most relevant predictors for COVID-19 diagnosis are sob severity, cough severity, sob presence, cough presence, fatigue and number of days since symptom onset. Originality/value Treatment for COVID-19 is not available to date. The best way to control this pandemic is the isolation of positive persons. It is very much necessary to identify positive persons at an early stage. RT-PCR test used to check COVID-19 positivity is the time-consuming, expensive and laborious method. Current diagnosis methods used in hospital demand more medical resources with increasing cases of coronavirus that introduce shortage of resources. The developed model provides solution to the problem cheaper and faster decreases the immediate need for medical resources and helps in decision-making.

TCT CONNECT-219 Psychosocial Impact of COVID - 19: Insights From a Cohort of Health Care Workers in the Cardiac Intensive Care Unit of a Tertiary Care Hospital in India

Background: COVID-19 has been the catalyst for a quantum shift in our professional and personal lives, literally and figuratively within the blink of an eyelash. Healthcare workers (HCWs) have been profoundly impacted by this disruption at all levels, especially those working in high-stress specialties, such as cardiology, in resource-deprived and population-dense areas in developing countries, such as India. We studied the impact of COVID-19 on a cohort of HCWs working in a high-stress, high-turnover cardiac intensive care unit (CICU) of a tertiary care center in India. Questionnaires, results, and conclusions detailed in this presentation. Considering the fact that India has not even reached the peak of the pandemic, the negative psychosocial impact of COVID-19 on HCWs of the cardiovascular community is highly concerning and disheartening. Simplistic, sustainable long-term action plans are the need of the hour. We must use the cataclysm wrought by COVID-19 to plug our broken healthcare systems. For that, our frontline warriors should be in the best state of physical, mental, and emotional well-being to face up to this challenge. The time to take action is NOW!! Methods: Evaluate the psychosocial impact of COVID-19 on HCWs working in a highly-stressed environment with high patient burden and turnover rates (45 bedded CICU including 15 step-down beds;average occupancy 90% to 100%). Understand perceived psychological burden and risk of post-traumatic stress disorder [PTSD] in these HCWs. [Formula presented] Delineating Stressors for this HCW Cohort A. Public Healthcare System and Bureaucratic Policies 1. Population-dense and resource-scarce developing Country 2. Despair at the inadequacy of the public healthcare system 3. Inability to understand and / or keep pace with fast-changing bureaucratic policies B. Updating knowledge on COVID-19 1. Stress of updating oneself on explosion of knowledge on COVID-19 2. Ambiguous / continuously-evolving admission guidelines and treatment algorithms on: i. COVID-19 patients with cardiac manifestations ii. COVID-19 patients presenting with an cardiac ailment iii. Patients presenting with decompensation of previous cardiac ailments due to COVID-19 Defining Stressors for this HCW Cohort C. Personal Protective Equipment [PPE] 1. Training on Donning and Doffing 2. Efficacy 3. Quality 4. Availability D. Workplace Disruption 1. Disruption of the existing workspace to incorporate demands of the pandemic 2. Redeployment to COVID-19 ICU 3. Creation of a separate cardiac ICU for COVID-19 cardiac patients 4. Revising process flows for admission / discharge / follow- up for cardiac patients E. Direct Psychological Impact of the Pandemic 1. Fear of getting infected by the virus 2. Fear of being an asymptomatic carrier for friends and family 3. Fear of job security 4. Jeopardy of future professional goals / growth / goals due to pandemic 5. Feelings of inadequacy / hopelessness / helplessness burn-out F. Societal Stigma / Ostracization due to Professional Hazard Snapshot - Questionnaire 1 to Assess Psychosocial Impact of COVID-19 * A. Personal Sphere: 1. Fear that you / your family are infected by COVID-19 2. Your likelihood of getting infected by COVID-19 3. Your fear regarding possibility of becoming infected by COVID-19 4. Increased familial responsibilities / load 5. Inability to destress as leisure and / or social activities impacted 6. Preventive measures at home to protect relatives during pandemic [ shift to another place - 10/ separate room - 6 / PPE at home while in common places – 4/ not possible to distance - 0] * Scale of 01 – 10 [01 – lowest / least;10 – highest / maximum] B. Professional Sphere – Impact of COVID-19 on Daily Workload 1. Tense working environment 2. Scarce social support from team 3. PPE usage a burden ( gloves, aprons, long-sleeved gowns, surgical masks, eye goggles) ? 4. Willingness to get redeployed to COVID-19 ICU, if required at peak of pandemic 5. Willingness to transit to remote / virtual consults for follow-up patients [post-discharge] 6. Increa ed bureaucracy 7. Extended working hours * Scale of 01 – 10 [01 – lowest / least;10 – highest / maximum] Snapshot - Questionnaire 2 to Assess Psychosocial Impact of COVID-19 ** 1. Emotionally drained from work 2. Fatigued in morning at prospect of going to work 3. Working with people all day – strain 4. Burnt – out / frustrated 5. Trouble sleeping 6. Reminder of negative work situations pertaining to pandemic 7. Irritable / angry 8. Surreal / unreal 9. Denial that in midst of a pandemic 10. Numb 11. Jumpy 12. Trouble with concentration ** Rate – 01 - Never / 02 - rarely / 03 - sometimes / 04 - frequently / 05 – always Snapshot - Questionnaire 3 to Assess Psychosocial Impact of COVID-19 *** 1. Feeling nervous, anxious or on the edge 2. Not able to stop / control worrying 3. Little interest / pleasure in doing things 4. Feeling down, depressed, or hopeless 5. Asked for psychological support 6. Heading for burnt – out syndrome 7. Afraid of suffering from burnout 8. Afraid of asking for help Timeframe: April – June 2020 *** Scale: 01 – Never / 02 – less than 10 days / 03 - less than 1/4th of three months / 04 – more than 50% of timeframe / 05 – Nearly everyday Results: Based on the preliminary data accumulated from this HCW cohort, COVID-19 has had a major negative psychosocial impact on it. 40% of HCWs are fearful of getting infected with COVID-19 and /or infecting family with it. 5% resigned from fear of contracting the virus. 35% faced social ostracization / discrimination to some extent since the pandemic onset. 3% came in contact with suspected COVID-19 patients and were quarantined, which has led to PTSD-like symptoms in that subset. Based on the preliminary data accumulated from this HCW cohort, COVID-19 has had a major negative psychosocial impact on it. 60% feel that their stress levels are higher by 50% or more due to direct / indirect impact of the pandemic. 70% opine that their workloads are higher since pandemic onset due to additional bureaucratic demands / PPE burden / reinforcement of infection control protocols / ambiguous admission / treatment criteria. 80% believe that they / their family members are at heightened risk of getting infected with COVID-19. 60% have symptoms of burnout. However, 80% of these hesitant to seek help for symptoms of burn-out Conclusion: Action Plan: Simplified admission / discharge / follow-up protocols Emphasis on virtual / remote follow-up visits for patients Education / awareness on COVID-19, especially for nursing staff Addressing PPE usage, its gaps and emphasizing adherence to it Addressing gaps in infection control protocol Reducing viral load by staggered workhours / breaks Anonymous in-house psychologist counseling options Categories Other: COVID-19 Lectures