Detecting COVID-19 Vaccine Stance and Symptom Reporting from Tweets using Contextual Embeddings

Determining sentiments of the public with regard to COVID-19 vaccines is crucial for nations to efficiently carry out vaccination drives and spread awareness. Hence, it is a field requiring accurate analysis and captures the interest of many researchers. Microblogs from social media websites such as Twitter sometimes contain colloquial expressions or terminology difficult to interpret making the task a challenging one. In this paper, we propose a method for multi-label text classification for the track of”Information Retrieval from Microblogs during Disasters (IRMiDis)” presented by the”Forum of Information Retrieval Evaluation” in 2022, related to vaccine sentiment among the public and reporting of someone experiencing COVID-19 symptoms. The following methodologies have been utilised: (i) Word2Vec and (ii) BERT, which uses contextual embedding rather than the fixed embedding used by conventional natural language models. For Task 1, the overall F1 score and Accuracy are 0.503 and 0.529, respectively, placing us fourth among all the teams, while for Task 2, they are 0.740 and 0.790, placing us second among all the teams who submitted their work. Our code is openly accessible through GitHub. 1 © 2022 Copyright for this paper by its authors.

Healthcare Seeking Behaviour of Hospitalised COVID-19 Patients During Second Wave in Tertiary Care Hospital of Northern India

Background: The Coronavirus Disease (COVID-19) pandemic continues its deadly reign all over the world. Devising effective strategies for detecting and controlling the infection has become ever more critical. Effective prevention and control of the pandemic is entirely dependent on human behavior in terms of practicing preventive and curative measures. During the second wave of COVID-19, people's perceptions of preventive and curative measures changed.Objective: To study healthcare-seeking behavior of hospitalized COVID-19 patients.Methods: Hospitalized patients due to COVID-19 in the month of March, April and May of 2021 were included in the study. Their attendants/close relatives were contacted telephonically to know about the admitted patients' healthcare-seeking behavior. Verbal consent was taken from attendants before the commencement of the interview, followed by informing them about the purpose of the interview.Results: Amongst the subjects, there were more males than females (67.5 vs 32.4%), age ranged between 18 to 88 with a mean value of 56.61 +/- 14.7 years. Self-medication was significantly associated with study subjects' mortality (p=0.03).Conclusion: Elderly people were having higher mortality rate than their younger counterparts. People were hesitant to visit primary care physicians after having symptoms of COVID-19.

ADAPTATION OF DOMAIN-SPECIFIC TRANSFORMER MODELS WITH TEXT OVERSAMPLING FOR SENTIMENT ANALYSIS OF SOCIAL MEDIA POSTS ON COVID-19 VACCINE

Covid-19 has spread across the world, and several vaccines have been developed to counter its surge. To identify the correct sentiments that are associated with the vaccines from social media posts, we fine-tune various state-of-the-art pretrained transformer models on tweets that are associated with Covid-19 vaccines. Specifically, we use the recently introduced state-of-the-art RoBERTa, XLNet, and BERT pre-trained transformer models, and the domain-specific CT-BERT and BERTweet transformer models that have been pre-trained on Covid-19 tweets. We further explore the option of text augmentation by oversampling using the language model-based oversampling technique (LMOTE) to improve the accuracies of these models – specifically, for small sample data sets where there is an imbalanced class distribution among the positive, negative, and neutral sentiment classes. Our results summarize our findings on the suitability of text oversampling for imbalanced small-sample data sets that are used to fine-tune state-of-the-art pre-trained transformer models as well as the utility of domain-specific transformer models for the classification task. © 2023 Author(s). This is an open access publication, which can be used, distributed and reproduced in any medium according to the Creative Commons CC-BY 4.0 License.

Predictive Analytics of Stock Market as a Time Series

This paper presents a new methodology and a comparative study using past stock market data that can help businesses take investing or divesting decisions in critical situations in the future. These may be like the COVID-19 pandemic, where market volatility is extremely high, thus creating an urgent need for better decision support systems to minimise loss and ensure better profits. The results of the study are based on the comparison of different configurations of ARIMAX, Prophet, LSTM and Bidirectional LSTM Models trained on historical NSE data. By understanding the correlation and variations in the data processing and model training parameters, we have successfully proposed a LSTM neural network model training and optimising method which could successfully help businesses take both long and short term profitable decisions before and after big financial and market crises with a respective accuracy of 98.60 percent and 96.97 percent. © 2022 IEEE.

Mepolizumab effectiveness in severe asthma supported by federated analysis of European SHARP data

Background: Innovative information techniques are increasingly used to perform federated analyses in real-world studies. Whether these techniques are suitable for harmonizing patient data from non-standardized registries and evaluating treatment outcomes needs further evidence. Aim(s): To standardize patient-level registry data from SHARP (Severe Heterogeneous Asthma Registry Patientcentred) and evaluate the effectiveness of mepolizumab on frequent (>=2/yr) exacerbations in patients with severe asthma. Method(s): We standardized data from 5,871 adults with severe asthma in 10 European countries using the OMOP Common Data Model (www.ohdsi.org). Patients who had taken mepolizumab >=1 yr (2016-2021) and had exacerbation data available were included. Changes in odds of >=2 exacerbations/yr were evaluated. Result(s): Of 2,109 patients who initiated mepolizumab 563 met inclusion criteria. Analysis showed a reduction of having >=2 (vs 0-1) annual exacerbations after 1 yr mepolizumab therapy: OR (95%CI) 0.18 (0.13-0.25)[N=369] pre and 0.08 (0.05-0.13)[N=194] during the COVID-19 pandemic (Fig). Conclusion(s): By harmonizing non-standardized, patient-level registry data and applying federated analysis we demonstrated that mepolizumab reduced asthma exacerbations, consistent with current knowledge. This paves the way for future pan-European real-world severe asthma studies using patient-level data in a privacy-proof way. (Figure Presented).

Big data and data mining in e-Health: Legal issues and challenges

e-Health is a rising industry. At a time when physical healthcare facilities were full of COVID19 patients, the e-Health Industry has become even more diverse and is being resorted to as a primary healthcare system specially to treat regular health problems. The health data of millions of patients is being stored online. The same is done through the concept of Big Data and Data Mining in e-Health. In India, the National Digital Health Mission is aimed towards using this technique to simplify e-Health services. Though Big Data helps in optimizing operations and services, there is another issue which is rising these days. The issue pertains to the Privacy of an individual's sensitive personal data and information. Health data, which is considered as the most sensitive data of the individual, is stored online and is made available for the use of various stake holders. In such a scenario, the issue of privacy becomes even more prominent, and the security of the data becomes even more important. This paper studies the use of Big Data in context of Data Privacy. The paper focuses on the legal framework with respect to privacy in India and a comparison of the same with other countries. © 2022 Scrivener Publishing LLC.

Googling "Anatomy dissection": Pre-pandemic, pandemic, and current scenario.

OBJECTIVES: The present study was proposed to assess if the online search behavior for Anatomy dissection was affected by actual dissection being performed in the labs. The interest in dissection was ascertained and compared between pre-pandemic, pandemic, and current times. METHODS: Online web search behavior for key terms related to "Anatomy dissection" was noticed worldwide using Google Trends Tool. Relative search volume (RSV) was downloaded for these keywords worldwide using all categories, and web search settings during a period from January 1, 2018, to July 31, 2022. RESULTS: There was almost a similar google trend for the first three months in years 2018 and 2020, with a slightly lower RSV in March and a noticeable RSV decrease from April onwards in 2020. During the pre-pandemic period, the peak for RSV in the month of May was noticed. These peaks were missing in 2020 and 2021. In May 2022, RSV again increased. The top five countries with the highest search hit for "Anatomy Dissection" web were the Philippines, the United States, Canada, Australia, and India. CONCLUSIONS: There was lower search trend for "Anatomy dissection" during the pandemic period as compared to pre-pandemic period. The search trends may be associated with the closure of dissection labs and it may be concluded that Anatomy learners search for this term more often, if they are actually performing the dissections. The future integrated multi-country data and analysis from different set-ups about their ways of dissection and labs before and after the pandemic may further clarify.

Analysis of Covid-19 Using Chest X-Ray Images: An AI Based Prospective

The unprecedented surge of Disease occurred in 2019, December named under covid-19, also known as corona-virus, was responsible for outbreaks happening in 2019-2020. It was first diagnosed in Wuhan, Hubei Province in the country China. Coronavirus is phenotypically and genotypically diverse in its behavior. Further, confirmed worldwide cases on April 6, 2020, consisting of 12,11,214 cases reported to the World Health Organization (WHO). The reverse-Transcript-Ase polymer-Ase chain reaction (RT-PCR) is the common solution utilized in the recognition of coronavirus, which has been found to be time-consuming and troublesome. On February 14, 2020, the laboratory-confirmed cases were 49,053 and deaths including 1,381 were documented. As of August 12, 2020, the virus has afflicted the lives of 20,674,903 people. It shows that deadly diseases are surging, plunging, and scattering illiberally in the current world. In accordance with the World Health Organization (WHO), over and above, 30 peculiar diseases like SARS coronavirus (SARS-COV), MERS coronavirus (MERS-COV), and H1N1 influenza, etc., have been diagnosed in previous twenty years to affect the health fitness situation of billions of people. The prime objective of this article is to make a review paper with artificial intelligence as a primary objective. It takes all the research to draw the comparative analyses of the distinct characteristics of data, mainly x-ray images of the chest, providing technologies and approaches to detect and prognosis coronavirus with learning algorithms. With the aim of finding the best-suited model, we compared the research on the basis of essential factors such as size of the batch , rate of learning, summed epochs till which it runs, and nature of optimizer. © 2022 IEEE.

Added value of the measles-rubella supplementary immunization activity in reaching unvaccinated and under-vaccinated children, a cross-sectional study in five Indian districts, 2018-20.

INTRODUCTION: Supplementary immunization activities (SIAs) aim to interrupt measles transmission by reaching susceptible children, including children who have not received the recommended two routine doses of MCV before the SIA. However, both strategies may miss the same children if vaccine doses are highly correlated. How well SIAs reach children missed by routine immunization is a key metric in assessing the added value of SIAs. METHODS: Children aged 9 months to younger than 5 years were enrolled in cross-sectional household serosurveys conducted in five districts in India following the 2017-2019 measles-rubella (MR) SIA. History of measles containing vaccine (MCV) through routine services or SIA was obtained from documents and verbal recall. Receipt of a first or second MCV dose during the SIA was categorized as "added value" of the SIA in reaching un- and under-vaccinated children. RESULTS: A total of 1,675 children were enrolled in these post-SIA surveys. The percentage of children receiving a 1st or 2nd dose through the SIA ranged from 12.8% in Thiruvananthapuram District to 48.6% in Dibrugarh District. Although the number of zero-dose children prior to the SIA was small in most sites, the proportion reached by the SIA ranged from 45.8% in Thiruvananthapuram District to 94.9% in Dibrugarh District. Fewer than 7% of children remained measles zero-dose after the MR SIA (range: 1.1-6.4%) compared to up to 28% before the SIA (range: 7.3-28.1%). DISCUSSION: We demonstrated the MR SIA provided considerable added value in terms of measles vaccination coverage, although there was variability across districts due to differences in routine and SIA coverage, and which children were reached by the SIA. Metrics evaluating the added value of an SIA can help to inform the design of vaccination strategies to better reach zero-dose or undervaccinated children.

Clinico-Laboratory Profile, Intensive Care Needs and Short-Term Outcome of Multisystem Inflammatory Syndrome in Children (Mis-C): Experience during First and Second Waves from India

BACKGROUND AND AIM: The multisystem inflammatory syndrome in children (MIS-C) is a new entity and needs data to study its evolution. To describe the clinicolaboratory profile, intensive care needs, and outcome of MIS-C during the first and second waves. METHOD(S): Retrospective analysis of 122 children with MIS-C admitted to Pediatric emergency and PICU of a tertiary-teaching hospital during first and second wave of Covid-19. RESULT(S): Median (IQR) age was 7 (4-10) years with 67% boys. Common manifestations included fever (99%), abdominal symptoms (81%), rash (66%), conjunctival injection (65%), oral mucosa and respiratory involvement (43% each). Elevated CRP (97%), D-dimer (89%), procalcitonin (80%), IL-6 (78%), ferritin (56%), NT-pro- BNP (84%), and positive SARS-CoV-2 antibody (81%) were common laboratory abnormalities. Cardiovascular manifestations included myocardial dysfunction (55%), shock (48%), and coronary artery changes (10%). The treatment included intensive care support (57%), non-invasive (33%) and invasive (18%) ventilation, vasoactive drugs (47%), IVIG (83%), steroids (85%), and aspirin (87%). Mortality was 5% (n=6). Duration of hospital stay was 5 (3-8) days. During second wave, significantly higher proportion had positive SARS-CoV-2 antibody, contact with COVID-19 case, and oral mucosal changes;lower markers of inflammation (CRP, procalcitonin, ferritin, and IL-6);lower rates of shock, myocardial dysfunction, and coronary artery changes;lesser need of PICU, vasoactive drugs, and IVIG;and shorter hospital stay. CONCLUSION(S): MIS-C is febrile multisystemic disease characterized by hyperinflammation, cardiovascular involvement, relationship to SARS-CoV-2, and good outcome with immunomodulation and intensive care. During the second wave, the severity of illness, degree of inflammation, and intensive care needs was lesser.

Prevalence and determinants of sexual abuse among adolescent girls during the COVID-19 lockdown and school closures in Ghana: A mixed method study.

BACKGROUND: In the wake of the COVID-19 pandemic, concerns have been raised that the pandemic may derail global efforts against child sexual abuse (CSA). OBJECTIVES: This study examines the prevalence and associated factors of sexual abuse among adolescent girls in the context of the COVID-19 pandemic in Ghana. PARTICIPANTS AND SETTING: The sample comprised 853 adolescent girls aged 13-19 (16.03 ± 2.04 years) in Ghana. METHODS: The study employed a concurrent mixed-method design. RESULTS: Overall, the prevalence of CSA during the COVID-19 lockdown and school closures was 32.5 %. Protective factors for CSA were feeling safe in neighbourhood (AOR = 0.526, 95 % CI = [0.325, 0.850]) and parents often listen to opinions (AOR = 0.446, 95 % CI = [0.241, 0.826]). Risk factors for CSA were physical activity (AOR = 1.649, OR = 1.783, 95 % CIAOR = [1.093, 2.487, 95 % CIOR = [1.241, 2.561]), parents sometimes listen to opinions (AOR = 1.199, OR = 1.924, 95 % CIAOR = [0.504, 2.853], 95 % CIOR = [1.034, 3.582]), living with another relative (AOR = 2.352, OR = 2.484, 95 % CIAOR = [0.270, 20.523], 95 % CIOR = [0.317, 19.475]), Akan ethnicity (AOR = 1.576, OR = 1.437, 95 % CIAOR = [0.307, 8.091], 95 % CIOR = [0.316, 6.534]), having no disability (AOR = 1.099, OR = 1.138, 95 % CIAOR = [0.679, 1.581], 95 % CIOR = [0.786, 1.649]) and having a close relationship with parents (AOR = 1.334, OR = 1.752, 95 % CIAOR = [0.746, 2.385], 95 % CIOR = [1.096, 2.802]). CONCLUSION: Knowledge of the risk and protective factors identified in this study can guide and inform the development of CSA prevention programmes during disruptive occurrences like school closures and lockdown.

Mortality Risk among Pregnant Women during COVID-19 Pandemic: A Case-Control Study in Rajasthan

Objective: To analyze the risk factors associated with mortality in COVID-positive pregnant women. Material(s) and Method(s): The study was designed as a case-control study and was conducted at COVID-designated tertiary care hospital. It included 42 deceased and 96 RT-PCR-positive surviving pregnant women who developed severe forms of disease. All the affected patients were admitted and managed according to Indian national guidelines. Risk factors - age, residence, socioeconomic status, gestation age, parity, and comorbidities were evaluated for their association with mortality. Result(s): Advanced maternal age (>28 years), rural residence, diabetes, and comorbidities were found to be associated with increased risk of mortality among the COVID-affected pregnant women, the respective adjusted odds ratio was 15.45, 12.61, 2.65, and 4.77, respectively. Conclusion(s): Pregnant women are vulnerable to COVID infection. They were at higher risk of mortality with comorbidities, advanced age, and low access to healthcare in the rural areas. Copyright © The Author(s).

Virtual NephroTalk: Teaching Communication Skills During the COVID-19 Pandemic

Background: Nephrologists engage in advance care planning with their patients less often than other specialists caring for patients with similarly life-limiting illnesses, and many patients with kidney disease receive invasive care that may not be consistent with their wishes. The UPMC NephroTalk curriculum has previously been shown to increase both self-rated preparedness and objective performance of essential communication skills to help prepare nephrology fellows to discuss goals of care with patients and families. The COVID-19 pandemic has necessitated innovation in teaching communication skills. We adapted the NephroTalk curriculum to provide the opportunity to acquire and practice communication skills while learning remotely. Method(s): Nephrology fellows from 4 ACGME-accredited nephrology programs, as well as other renal healthcare professionals participated in a three-day virtual course with synchronous and asynchronous components including 1) self-paced modules explaining and demonstrating communication skills;2) small group drills to promote deliberate practice of core skills led by experienced facilitators;and 3) small group practice with simulated patients incorporating peer and facilitator feedback. Surveys measured participants' self-perceived communication preparedness pre- and post-curriculum and satisfaction with the program. Result(s): Twenty-two healthcare professionals participated in the virtual NephroTalk curriculum over two years. Prior to the course, most learners described feeling "not prepared" or "somewhat prepared" on a 5-point Likert scale to perform 12 key communication skills including discussing prognosis, eliciting patient values, discussing conservative kidney management, and recommending a treatment plan. After completing the program, 100% of participants reported feeling "prepared" or "very prepared" to perform all 12 communication skills. Respondents expressed a high degree of satisfaction with the course overall. Conclusion(s): The COVID-19 pandemic has resulted in a shift toward virtual learning within medical education. Nephrology fellows and clinicians participating in the virtual NephroTalk curriculum expressed high levels of satisfaction and preparedness to utilize the skills demonstrated in the course. Future research should evaluate the impact of virtual communication skills training on sustained skill maintenance over time and patient outcomes.

To Kayexalate or Not to Kayexalate

Introduction: Although rare, it is important to consider colon necrosis, ulceration, and perforation in the management of patients with abdominal pain following kayexalate administration. We present a critically ill patient requiring partial hemicolectomy following kayexalate for hyperkalemia. Case Description: A 43-year-old male presented to the ER via EMS after being intubated emergently in the field. Was found to be COVID-19 positive and admitted to the ICU. Creatinine and potassium started trending up after day 5, requiring multiple doses of calcium gluconate, insulin, and dextrose, along with 4 doses of kayexalate 30 g over the second week. On the 10th day, he started spiking fever, having abdominal distension, and continued desaturating. Abdominal x-ray showed a large amount of free intraperitoneal air warranting emergent ex lap with partial right hemicolectomy and end ileostomy. Unfortunately, he was pronounced dead after a lengthy hospital stay of 55 days. The specimen showed perforation, mucosal necrosis, and acute serositis. It also showed Amphophilic crystals suggesting a diagnosis of kayexalate-induced colon ischemia and necrosis. Discussion(s): Kayexalate was approved by the FDA in 1958 and has been used to treat hyperkalemia. It can bind intraluminal calcium, leading to bowel obstruction or perforation, with a reported incidence of 0.14-1.8%. The identification of rhomboid or triangular, basophilic crystals with a mosaic pattern on H&E stain is pathognomonic for the presence of kayexalate. We present this case as a reminder of the rare yet devastating complications of kayexalate. For that reason, clinical suspicion should be raised in patients with abdominal pain following kayexalate. Kayexalate should only be used in patients who have life-threatening hyperkalemia where dialysis or newer cation exchangers (ie, patiromer or Lokelma) are not available, and other therapies to remove potassium have failed or are not possible. (Figure Presented).

Measurement of Optic Nerve Sheath Diameter by Trans-Orbital Ultrasound to Detect Raised Intracranial Pressure in Children

Aims Raised Intracranial Pressure (ICP) is frequently encountered condition amongst children due to neurological and non-neurological etiologies. It contributes towards 20% of the admissions in our PICU at tertiary care institute of North India. Timely diagnosis of raised intracranial pressure is critical for appropriate management. Ultrasonographic measurement of optic nerve sheath diameter (ONSD) is non-invasive point of care tool to recognize raised intracranial pressure however there are very limited studies in pediatric population on this. Aim of this study was to measure the ONSD of children admitted in pediatric intensive care unit (PICU) to identify the difference in ONSD among the pediatric patients with normal and raised ICP and to achieve the cut-off value for diagnosing raised ICP. Methods A Hospital based observational comparative study conducted at PICU of a tertiary care institute of national importance in North India. ONSD measurement of all the patients aged 2-14 years admitted in pediatrics department for >48 hours was done in two phases due to COVID-19 related restrictions in routine admissions for some months. First phase was from 1st January 2020 to 31st March 2021 and second phase started from 15th January 2022. Complete enumeration technique was used for the study. Children receiving treatment for raised ICP based on clinical features (Muir's criteria) and neuroimaging were included in raised ICP group and they were compared with the patients without any signs of raised ICP. Optic nerve sheath diameter was measured in both the groups on Day 1 (i.e. within 24 hours of admission) and Day 2 of admission. Third and last ONSD measurement was done on any day between day 4 and day 7 of admission. On each day, 3 readings were taken from each eye and average of 3 readings was used for final statistical analysis for that particular day. Treating team was unaware of the USG findings throughout the stay of the patient in the department. Results Total 101 patients recruited, among them 16 patients had raised ICP and 85 patients had normal ICP. The mean optic nerve sheath diameter on day 1 was found to be higher in the patients with raised ICP (Mean ONSD = 4.955 +/- 0.80) as compared to those with normal ICP (Mean ONSD = 4.05 +/- 0.43). Their mean difference was 0.90 mm (95% CI 0.07-1.48 p value <0.01). Mean ONSD on day 2 also was higher in raised ICP patients (Mean ONSD = 4.82 +/- 0.73) in comparison to normal ICP patients (Mean ONSD = 4.06 +/- 0.47) which was statistically significant (p value = 0.001). The cut-off ONSD value for detecting the raised ICP was estimated to be 4.35 mm on ROC curve with area under the curve 0.862 (95% CI, 0.721-0.943), sensitivity of 75% and specificity of 76.5%. Conclusion Raised ICP is one of the important neurological entities in children contributing towards significant morbidity. Measurement of ONSD by trans-orbital ultrasound was able to detect raised ICP with excellent discriminatory performance.

Impact of Covid-19 Pandemic on Teaching and Training of Orthopaedic Resident in Medical College

Background: Many elective surgeries are getting postponed which directly affect the learning of surgical skill in resident. Various teaching activity such as OPD and bed side teaching also reduced because of decrease load and social distancing. Purpose of the study is to analyse the impact of this pandemic on orthopaedic education of resident in medical college. Material and Methods: A questionnaire was prepared on Google form for online survey. Total 74 residents responded, in which JR-1 (26), JR2 (24), JR-3 (19) and SR (5) participated. This study is an observational study in which we analysed the impact of this pandemic on orthopaedic teaching and training of resident in medical college. Result: In this pandemic clinical work in the form of OPD, IPD, no. of surgery (both elective and emergency) observed, assisted, independently performed significantly reduced. This directly affected the skill development of growing orthopaedic surgeon. Classroom teaching and bed side teaching were grossly affected. Some online platforms were used to teach the resident, but they are not as effective as conventional methods. Because of decrease clinical workload resident’s thesis and research work also affected which is an essential part of resident education. Conclusion: In summary, the results of this online survey on residents showed that there significant impact of the COVID-19 pandemic on teaching and training of orthopaedic resident in government medical colleges of central India.

Aspect-Based Sentiment Analysis of online Reviews for Business Intelligence

The COVID-19 pandemic has affected the patients to a large extent who are in immediate need of medical care. People rely on online reviews shared on review websites to gather information about hospitals like the availability of beds, availability of ventilators, etc. However, these reviews are large in number and unstructured, which makes it difficult to interpret them. Hence, the authors have proposed a methodology that will apply an aspect-based sentiment analysis on the reviews to gain meaningful insights of the hospital based on different aspects like doctor, staff, facilities, etc. For the purpose of empirical validation, a total of 26,071 reviews pertaining to 325 hospitals were scrapped. The authors concluded that the study can be useful for patients as it helps them to select the hospital that best suits them. The study also helps hospital administration to improve the current services according to the needs of the patients. Copyright © 2022, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.