Analysis of Communicable Disease Symptoms Using Bag-of-Neural Network at Edge Networks

Considering the increasing growth of communicable diseases worldwide such as COVID-19, it is recommended to stay at home for patients with fewer chronic health problems. In recent times, the high chance of COVID-19 spread and the lack of an excellent remote monitoring system make the situation challenging for hospital administrators. Inspired by these challenges, in this paper, we develop a new edge-centric healthcare framework for remote health monitoring and disease prediction using Wearable Sensors (WSs) and advanced Machine Learning (ML) model, namely Bag-of-Neural Network (BoNN), respectively. The epidemic model collects the health symptoms of the patient using various a set of WSs and preprocesses the data in distributed edge devices for preparing a useful dataset. Finally, the proposed BoNN model is applied over the refined dataset for detecting COVID-19 disease at centralized cloud servers using a set of random neural networks. To demonstrate the efficiency of the proposed BoNN model over the standard ML models, the system is fine-tuned and trained over a synthetic COVID-19 dataset before being evaluated on a benchmark Brazil COVID-19 dataset using various performance metrics. The experimental results demonstrate that the proposed BoNN model achieves 99.8% accuracy while analyzing the Brazil dataset. © 2001-2012 IEEE.

Wavelet-based Spectral Analysis For Protein Conformation Selection and Prediction Using AI in Drug Discovery Applications

Contemporary drug discovery relies heavily on massive high performance computing (HPC) resources from docking and molecular dynamics simulations of proteins interacting with drug candidate ligands, such as recently published form simulations on ORNL's SUMMIT in covid19 pharmaceutical research. This work presents a unique spectral analysis approach using wavelet transform (WT) to understand the correlation between the time evolution of protein conformations generated by molecular dynamics and specific protein conformations that are selected for binding by ligands. A DWT-based spectral analysis is performed on the unique protein descriptors previously identified to be important in protein: ligand binding. The new protein time-series information from the wavelet-based time-frequency domain analysis is used for a more refined protein conformation selection and improve the deep learning and machine learning (AI/ML) prediction framework to improve the prediction of binding vs. non-binding protein conformations for three target proteins ADORA2A, OPRD1 and OPRK1. In this work, wavelets are used for spectral analysis for their added benefit of simultaneous time-frequency resolution and denoising properties. © 2022 IEEE.

Abracadabra Alakazam: Sudden Invasion of the Pericardium by Haemophilus Influenzae

A 25 year old man presented with three days of cough, shortness of breath, and pleuritic chest pain. Initial vital signs in the ED were normal, and exam demonstrated tonsillar erythema without exudate. Labs revealed a leukocytosis of 18.9k/muL, D-dimer of 690 ng/mL, C-reactive protein of 5.7 mg/dL, and lactate elevated to 2.9 mmol/L. High-sensitivity troponin, NT-proBNP, and SARS-CoV-2 RT-PCR were all negative. Presenting electrocardiogram demonstrated PR elevation in aVR with diffuse ST-segment elevation in the inferior and anterolateral leads. Point-of-care echocardiogram demonstrated normal biventricular function without pericardial effusion. CTPA was negative for pulmonary embolism, and he was observed for presumed acute viral pericarditis. Fourteen-hours later, he became febrile to 38.3degreeC, tachycardic to 133 bpm, and hypotensive to 97/65 mmHg with diffuse abdominal pain. Repeat lactate was 9.0 mmol/L. This prompted an emergent CT scan which now showed a new large pericardial effusion and bilateral pleural effusions (Panel A). Repeat echocardiogram confirmed a large circumferential pericardial effusion with early signs of tamponade including right atrial inversion in late diastole (Panel B). Emergent pericardiocentesis yielded 560 mL of brown, purulent fluid (Panel C) with immediate improvement in hemodynamics. Bacterial gram stain and culture grew Haemophilus influenzae (Panel D). Immunodeficiency screening was negative. Transient severe biventricular systolic dysfunction was noted, consistent with sepsis-induced cardiomyopathy. He completed a targeted antibiotic course with partial recovery of his ejection fraction by discharge. Purulent pericarditis is rare in developed countries, and invasive H. influenzae in a young, immunocompetent adult is particularly unusual. This case illustrates the importance of early diagnosis and management of purulent pericarditis given its potential for rapid progression and high mortality. (Figure Presented).

A Study on COVID–19, Its Origin, Phenomenon, Variants, and IoT-Based Framework to Detect the Presence of Coronavirus

An ongoing pandemic SS-RNA viral infection initiated from the Chinese province has threatened people throughout the globe. Coronavirus or COVID-19 or 2019-nCoV as a contagious infection is spreading day-by-day threatening the livelihood of people. The main objective of this paper is to find out solutions for the detection of this contagious viral infection at the earliest. Computer-based artificial intelligence can be used to monitor and detect the symptoms of coronavirus. For detection of coronavirus infection, computers or smartphones can be embedded with biosensors that will perceive the information and will convert the information into digital data. In this paper, a study on the coronavirus is done and an IoT-based framework is proposed to detect the coronavirus using IoT-based sensors. The proposed approach will be able to detect the pandemic in its early stages, and better options for prevention and cure will be discussed. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Patient Attitudes Toward Telemedicine in Head and Neck Oncology

Introduction: Despite dramatic expansion of telehealth visits across health disciplines during the COVID-19 pandemic, little is known about attitudes toward telehealth offerings among patients affected by head and neck neoplasms. We investigate patient responses regarding their desire for telemedicine, personal attitudes, and barriers in availing telemedicine visits for head and neck oncology-related care. Method(s): Retrospective analyses of patient surveys prospectively administered between August and October 2021 to adult patients at a tertiary care head and neck oncology clinic. Level of interest in telemedicine appointments was the assessed primary outcome. Covariates including demographics, place of residence, primary neoplastic condition, speech/communication barriers, access to internet-enabled devices or reliable internet, personal preferences, and qualitative self-reporting of attitudes toward telemedicine were assessed for association with interest in telemedicine appointments. Result(s): Of 633 survey responses, 50.6% were male. More than 70% of surveyed patients were older than 56 years. Overall, 49.7% demonstrated interest in telehealth visits. Limitations in access to technology (17.7% [112 of 633 respondents]) and lack of reliable internet connection (13.74% [87 of 633 respondents]) were considered key barriers. Only 6% patients expressed concerns about missing key elements of care or quality of care delivered through telemedicine visits. Conclusion(s): Limited access to technology platforms and unreliable internet were key concerns for 1 in 2 patients considering telemedicine appointments for head and neck oncologic care. Understanding the needs and attitudes of specific patient populations may be important for organizations pivoting to telemedicine platforms for improving health care access. Key interventions to enhance participation in telemedicine- based care delivery could include identifying rural connectivity hubs and ensuring availability of connected devices through grant or device loan programs, and employing userfriendly technology platforms.

Renal Function Dynamics Among Mechanically Ventilated Patients With Acute Kidney Injury (AKI) And COVID-19 Pneumonia

The incidence of COVID-19 associated AKI in hospitalized patients is variable, but majority of literature suggests that it affects >20% of the population. Long-term outcomes are also variable – a proportion of which may progress to chronic kidney disease (CKD) with worsening of baseline renal function. There is an insufficient data regarding long-term outcomes of AKI in intubated COVID-19 patients. The purpose of this study was to determine the likelihood and the risk of CKD in intubated COVID- 19 patients 90 days after the first episode of inpatient AKI. The study population included all intubated patients with confirmed COVID-19 pneumonia and bacterial pneumonia admitted from Jan 1, 2020 to Dec 31, 2020, who developed AKI during admission and were discharged alive. Diagnosis of AKI was based on KDIGO definition. Serum creatinine on admission, during first episode of AKI, on discharge and 90 days after first episode of AKI were collected. Logistic regression analysis was conducted to determine the 0dds ratio for CKD on day 90. Cox proportional hazard ratio was conducted to assess the risk of CKD at day 90 after AKI. The study included 125 patients. 56 (45%) had COVID-19 pneumonia while 69 (55%) had bacterial pneumonia. There was no noted differences in the baseline characteristics. Emergent inpatient hemodialysis was higher amongst COVID-19 patients (20% vs 7%, p<0.043), which equated to more patients requiring dialysis after discharge (18% vs 3%, p<0.005). The likelihood of having CKD in COVID-19 patients was 2.62 (1.06-6.45, p =0.037) and Hazard Ratio for CKD was 2.48 (1.06-5.78, p=0.036). COVID-19 patients with AKI has higher likelihood (2.62) and increased risk (148%) of developing CKD after AKI compared to bacterial pneumonia patients, regardless of comorbidities, mechanical ventilation days, or highest AKI stage. Critically ill COVID-19 patients requiring mechanical ventilator who develop AKI on admission need close monitoring of renal function both during hospital stay and most especially upon discharge. (Figure Presented)

Analysis of Communicable Disease Symptoms using Bag-of-Neural Network at Edge Networks

Considering the increasing growth of communicable diseases worldwide such as COVID-19, it is recommended to stay at home for patients with fewer chronic health problems. In recent times, the high chance of COVID-19 spread and the lack of an excellent remote monitoring system make the situation challenging for hospital administrators. Inspired by these challenges, in this paper, we develop a new edge-centric healthcare framework for remote health monitoring and disease prediction using Wearable Sensors (WSs) and advanced Machine Learning (ML) model, namely Bag-of-Neural Network (BoNN), respectively. The epidemic model collects the health symptoms of the patient using various a set of WSs and preprocesses the data in distributed edge devices for preparing a useful dataset. Finally, the proposed BoNN model is applied over the refined dataset for detecting COVID-19 disease at centralized cloud servers using a set of random neural networks. To demonstrate the efficiency of the proposed BoNN model over the standard ML models, the system is fine-tuned and trained over a synthetic COVID-19 dataset before being evaluated on a benchmark Brazil COVID-19 dataset using various performance metrics. The experimental results demonstrate that the proposed BoNN model achieves 99.8% accuracy while analyzing the Brazil dataset. IEEE

Clinical Efficacy of Inhaled Nitric Oxide in Preventing the Progression of Moderate to Severe COVID-19 and Its Correlation to Viral Clearance: Results of a Pilot Study

Hypoxic patients with coronavirus disease 2019 (COVID-19) are at high risk of adverse outcomes. Inhaled nitric oxide (iNO) has shown anti-viral and immunomodulatory effects in vitro. However, in vivo evidence of efficacy in hypoxic COVID-19 is sparse. This open label feasibility study was conducted at a single referral center in South India and evaluated the effectiveness of repurposed iNO in improving clinical outcomes in COVID-19 and its correlation with viral clearance. We recruited hypoxemic COVID-19 patients and allocated them into treatment (iNO) and control groups (1:1). Viral clearance on day 5 favored the treatment group (100% vs 72%, P < 0.01). The speed of viral clearance as adjudged by normalized longitudinal cycle threshold (Ct) values was positively impacted in the treatment group. The proportion of patients who attained clinical improvement, defined as a ≥2-point change on the World Health Organization ordinal scale, was higher in the iNO cohort (n = 11, 79%) as compared to the control group (n = 4, 36%) (odds ratio 6.42, 95% confidence interval 1.09-37.73, P = 0.032). The proportion of patients progressing to mechanical ventilation in the control group (4/11) was significantly higher than in the treatment group (0/14). The all-cause 28-day mortality was significantly different among the study arms, with 36% (4/11) of the patients dying in the control group while none died in the treatment group. The numbers needed to treat to prevent an additional poor outcome of death was estimated to be 2.8. Our study demonstrates the putative role of repurposed iNO in hypoxemic COVID-19 patients and calls for extended validation. Copyright © 2021 the Author(s). Published by Wolters Kluwer Health, Inc.

A Secured Real-Time IoMT Application for Monitoring Isolated COVID-19 Patients using Edge Computing

Internet of Medical Things (IoMT) is an emerging technology whose capabilities to self-organize itself on-the-fly, to monitor the patient's vital health data without any manual entry and assist early human intervention gave birth to smart healthcare applications. The smart applications can be used to remotely monitor isolated patients during this COVID-19 pandemic. Remote patient monitoring provides an opportunity for COVID-19 patients to have vital signs and other indicators recorded regularly and inexpensively to provide rapid and early warning of conditions that require medical attention using secured edge and cloud computing. However, to gain the confidence of the users over these applications, the performance of healthcare applications should be evaluated in real-time. Our real-time implementation of IoMT based remote monitoring application using edge and cloud computing, along with empirical evaluation, show that COVID-19 patients can be monitored effectively not only with mobility but also helps the health care professionals to generate consolidated health data of the patient that can guide them to obtain medical attention. © 2021 IEEE.

STRATEGIES TO CONTAIN COVID-19 INFECTION IN A HOSPITAL BASED OUTPATIENT HEMODIALYSIS UNIT IN THE SOUTH BRONX

New York City was the epicenter of COVID-19 infections within the United States in the spring of 20201. Our public, hospital-based hemodialysis (HD) unit is located in Bronx County, which had the highest rates of infections and deaths due to COVID-19.2 We retrospectively investigated the prevalence of COVID-19 in our HD unit and the effectiveness of expanded infection control measures implemented during the surge. Charts were reviewed for all 61 patients receiving maintenance HD between March 1-July 15, 2020. 4 HD patients and 2 HD healthcare providers (HCP) developed symptoms from COVID-19 infection between March 17-23, followed by another 5 patients and 2 HCP. HD patients underwent SARS-CoV-2 PCR nasal swab, regardless of symptoms, allowing detection of 4 asymptomatic COVID-19 cases. Positive cases were cohorted. Patients were screened for fever and COVID-19 symptoms before each HD, advised to wear face masks and practice hand hygiene. 5 patients were hospitalized with COVID-19 within 14 days of the screening period with no additional cases detected afterwards. During the surge, patients requiring bedside HD increased exponentially so HD frequency or treatment hours were reduced for some patients and 20 were temporarily transferred to other units. In May, all 32 HCP were tested for COVID-19 antibody with 18.8% (5 with and 1 without symptoms) testing positive. In June, 51 HD patients were tested for antibodies with detection of 6 additional asymptomatic individuals who had been SARS-CoV-2 PCR negative. In total, 26 patients (42.6%) tested positive for COVID-19, of which 42.3% were asymptomatic, and with 1 death. Early identification and isolation of both symptomatic and asymptomatic patients by universal screening along with stringent infection control measures limited the spread of COVID-19 infection in our unit.

A NOVEL APPROACH TO TEACHING CORONARY ANGIOGRAPHY USING 3D-PRINTING

Background: Coronary angiography (CAG) is a fundamental component of cardiology fellowship. As the impact of COVID-19 fueled the need for self-directed and remote learning, we sought to develop a resource that would address this need and improve angiogram interpretation skills among fellows. To this aim we developed a teaching tool correlating fluoroscopic projections with a 3D-printed physical and digital model of coronary anatomy derived from coronary CT. We hypothesized that fellows exposed to this resource would benefit from improved comprehension of spatial concepts in CAG compared to usual resources. Methods: Twenty-two cardiology fellows were randomly assigned to exposure of the teaching tool versus usual resources. An exam assessing comprehension of the spatial orientation of coronary anatomy, aortic cusps, and catheter tip position in relation to fluoroscopic views was administered before and after a six-week exposure period. Scores were compared, and qualitative feedback was obtained using the Likert scale. Results: Fellows exposed to the content achieved a greater improvement on their exam score and were more likely to improve (Figure). All fellows felt the content was superior to existing resources and will improve their comprehension of CAG. Conclusion: Educational platforms leveraging 3D printing can enhance comprehension of CAG among cardiology fellows, and may serve as valuable resources to promote self-directed and remote learning. [Formula presented]

Comparison of Bloodstream Infections between Hospitalized Patients with and without COVID-19 Infection during the First Wave of the COVID-19 Pandemic in a Community Hospital in South Bronx: An Observational Study

Background. Comparative data on bloodstream infections (BSI) in hospitalized patients with and without SARS-CoV2 positive test is lacking. Methods. A retrospective observational study comparing (BSI) with and without COVID-19 infection was performed was performed from Jan1- May 1, 2020. Patient demographics, clinical microbiological characteristics of infections, therapeutic interventions and outcomes was compared between the two groups. Results. Of 155 patients with BSI, 104 were SARS-CoV2 PCR negative (N) while 51 were positive (Table 1). Majority of SARS-CoV2 positives (P) had ARDS (58.8%), required mechanical ventilation (73%), inotropic support (55%), therapeutic anticoagulation (28%), proning (35%), Rectal tube (43%), Tocilizumab (18%), and steroids (43%) (Table 2). BSI was higher in N with HIV (16.3% vs 3.9% p=0.027). Duration of antibiotic therapy (DOT) prior to BSI was significantly longer in P (15 days vs. 5 days, p < 0.0001) (table 2). In-hospital mortality was significantly higher among P with BSI (49% vs. 21% p < 0.0001). 185 BSI events were observed during the study period with 117 in N patients and 68 in P. Primary BSI was predominant (76%) in N while secondary BSI (65%) was common in P of which 50% were CLABSI. Median time from admission to positive culture was 0.86 days in N compared to 12.4 in P (p = 0.001). Majority of BSI in P were monomicrobial (88%) and hospital acquired (71%) when compared to N (p< 0.001). Enterococcus spp (28%), Candida spp(12%), MRSA (10%) and E.coli (10%) were predominant microbes in P compared to Streptococcus grp (16%), MSSA (14%), MRSA (13%) and E.coli (12%) in N (figure 1). Mortality from BSI was associated with COVID-19 infection (OR 2.403, p = 0.038), DM (OR 2.335, p = 0.032), Charlson comorbidity index >3 (OR 1.236, p = 0.004), and mechanical ventilation (OR 11.398, p < 0.001) on multivariate analysis. Conclusion. Increased events of hospital acquired, secondary BSI (CLABSI) due to Enterococcus was observed in adult P compared to N. These patients were critically ill, developed BSI in the second week of hospitalization, had longer DOT prior to positive cultures and worse outcomes. Breakdown of infection control measures and inappropriate antimicrobial use during the surge could be contributory.

Comparison of Bloodstream Infections in Hospitalized Patients before and during the COVID-19 Surge in a Community Hospital in the South Bronx: An Observational Study

Background. There is a paucity of data of bloodstream infections (BSI) before and during the COVID-19 pandemic. The aim of our study was to compare the incidence and characteristics of blood stream infections (BSI) in hospitalized patients before and during the surge of COVID-19 pandemic in a community hospital in South Bronx. Methods. This is a retrospective observational comparative study of adult hospitalized patients with BSI admitted before (Jan 1-Feb 28, 2020) and during COVID-19 surge (Mar 1- May 1,2020). The incidence of BSI, patient demographics, clinical and microbiological characteristics of infections including treatment and outcomes were compared. Results. Of the 155 patients with BSI, 64 were before COVID and 91 were during the COVID surge (Table 1). Incidence of BSI was 5.84 before COVID and 6.57 during surge (p = 0.004). Majority of patients during COVID period had ARDS (39.6%), required mechanical ventilation (57%), inotropic support (46.2%), therapeutic anticoagulation (24.2%), proning (22%), rectal tube (28.6%), Tocilizumab (9.9%), and steroids (30.8%) in comparison to pre-COVID (Table 2). Days of antibiotic therapy prior to BSI was 5 days before COVID and 7 during COVID. Mortality was higher among patients with BSI admitted during COVID surge (41.8% vs. 14.1% p < 0.0001). Of 185 BSI events, 71 were Pre-COVID and 114 during surge. Primary BSI were predominant (72%) before COVID contrary to secondary BSI (46%) (CLABSI) during COVID. Time from admission to positive culture was 2.5 days during COVID compared to 0.9 pre-COVID. Majority of BSI during COVID period were monomicrobial (93%) and hospital acquired (50%) (p=0.001). Enterococcus (20.2%), E.coli (13.2%), and MSSA (12.3%) were predominant microbes causing BSI during COVID vs. MRSA (15.5%), Streptococci (15.5%), and S. pneumoniae (14.1%) before COVID (Figure 1). In multivariate logistic regression, Enterococcal coinfection was associated with COVID positivity (OR 2.685, p = 0.038), mechanical ventilation (OR 8.739, p = 0.002), and presence of COPD/Asthma (OR 2.823, p = 0.035). Conclusion. Higher incidence of secondary BSI (CLABSI) due to Enterococcus spp. was observed during the surge of COVID-19 infection in the South Bronx. Breakdown of infection control measures during the COVID-19 pandemic could have been contributory.

Persisting COVID-19 vaccination hesitancy in the South Bronx

Background. Minority groups have the lowest vaccination rates when compared to the overall population. We aim to study the attitudes and perceptions of COVID-19 vaccination, about six months after vaccine rollout in the South Bronx. Methods. Cross-sectional anonymized online survey evaluating knowledge, attitude and perception about COVID-19 vaccination using SurveyMonkey™ was conducted in South Bronx community from April - June 2021. Results. Of the 281 participants, 67% were Latinx and 16% were African American (AA);69% (195) were fully vaccinated (FV) and 31% (86) with vaccine hesitancy (VH). The common reasons for hesitancy were "concerns about side effects" (38%), "vaccine is not safe" (27%) and "vaccine was approved too fast" (26%) (p< .001). VH were more likely to rely online/mobile apps (30%) and friends and family (23%) as compared to FV. VH were more likely to be AA, younger age (< 35 yrs), high school or lower education, single, unemployed, without comorbidities, not current on other eligible vaccines, and did not believe "vaccine is necessary to end the pandemic." Majority of participants from both cohorts trusted their primary care providers. Mistrust with healthcare and pharmaceutical companies was higher in VH (p=0.009). Both groups preferred to continue wearing mask and practice social distancing despite vaccination status. Conclusion. Persisting vaccine hesitancy is concerning in minority communities. Identifying the target population and implementation of innovative methods to improve COVID-19 vaccination acceptance leveraging primary care providers would be a possible solution.

Casirivimab and Imdevimab for Treatment of Hospitalized Patients with COVID-19 Receiving Low Flow or No Supplemental Oxygen

Background. Casirivimab and imdevimab (CAS/IMDEV) is authorized for emergency use in the US for outpatients with COVID-19. We present results from patient cohorts receiving low flow or no supplemental oxygen at baseline from a phase 1/2/3, randomized, double-blinded, placebo (PBO)-controlled trial of CAS/IMDEV in hospitalized patients (pts) with COVID-19. Methods. Hospitalized COVID-19 pts were randomized 1:1:1 to 2.4 g or 8.0 g of IV CAS/IMDEV (co-administered) or PBO. Primary endpoints were time-weighted average (TWA) change in viral load from baseline (Day 1) to Day 7;proportion of pts who died or went on mechanical ventilation (MV) through Day 29. Safety was evaluated through Day 57. The study was terminated early due to low enrollment (no safety concerns). Results. Analysis was performed in pooled cohorts (low flow or no supplemental oxygen) as well as combined treatment doses (2.4 g and 8.0 g). The prespecified primary virologic analysis was in seronegative (seroneg) pts (combined dose group n=360;PBO n=160), where treatment with CAS/IMDEV led to a significant reduction in viral load from Day 1-7 (TWA change: LS mean (SE): -0.28 (0.12);95% CI: -0.51, -0.05;P=0.0172;Fig. 1). The primary clinical analysis had a strong positive trend, though it did not reach statistical significance (P=0.2048), and 4/6 clinical endpoints prespecified for hypothesis testing were nominally significant (Table 1). In seroneg pts, there was a 47.0% relative risk reduction (RRR) in the proportion of pts who died or went on MV from Day 1-29 (10.3% treated vs 19.4% PBO;nominal P=0.0061;Fig. 2). There was a 55.6% (6.7% treated vs 15.0% PBO;nominal P=0.0032) and 35.9% (7.3% treated vs 11.5% PBO;nominal P=0.0178) RRR in the prespecified secondary endpoint of mortality by Day 29 in seroneg pts and the overall population, respectively (Fig. 2). No harm was seen in seropositive patients, and no safety events of concern were identified. Conclusion. Co-administration of CAS/IMDEV led to a significant reduction in viral load in hospitalized, seroneg pts requiring low flow or no supplemental oxygen. In seroneg pts and the overall population, treatment also demonstrated clinically meaningful, nominally significant reductions in 28-day mortality and proportion of pts dying or requiring MV.