TLU-Net: Transfer Learning Framework using U-Net Convolutional Neural Networks for CT-based Lungs and COVID-19 Segmentation

Medical imaging has been a crucial component of COVID-19 clinical procedures. However, CT scans for lung and COVID-19 infection diagnosis are time-consuming and require specific knowledge. In addition, diverse imaging modalities, such as regions of interest (organs or lesions) and picture quality, make the automated segmentation of these infectious fragments difficult. In this paper, we proposed TLU-Net based on an optimized design of U-Net architectures to facilitate successful transfer learning for the segmentation of medical images from a source domain, such as lungs, to a target domain, such as COVID-19 infection. We have investigated the effective transfer learning methodologies for U-Net topologies, hyperparameters, freezing, and multiple models/datasets in the lungs and COVID-19 domains. Due to our improved and lightweight U-Net design, experiments with TLU-Net employing three lung and COVID-19 infection domain datasets increased segmentation accuracy and reduced training time. © 2022 IEEE.

Role of Tocilizumab in Patients with Moderate-to-Severe Covid-19 Pneumonia

Objectives: Cytokine release syndrome (CRS) is believed to be responsible for death in COVID-19. Tocilizumab is an interleukin (IL)-6 receptor antagonist, IL-6 being identified as a major component of the CRS cascade. The objective of the study was to determine if tocilizumab can prevent mortality and morbidity in moderate-to-severe COVID-19 pneumonia. Method(s): Patients admitted to the ICU between the time period of June 2020-August 2020 were included in this retrospective and cohort study conducted at GCS medical college, hospital and research center. Patients had to be more than 18 years of age and were required to have a positive reverse transcription polymerase chain reaction report for COVID-19. After applying the inclusion/exclusion criteria, 119 patients were considered for final analysis. Tocilizumab was administered as a single dose of 8 mg/kg in 22 patients. Rest of the patients received standard of care regime. The primary outcome was either discharge or death of the patients and the requirement of invasive mechanical ventilation during their hospital stay. The secondary outcome was the length of hospital stay. Appropriate demographic, clinical, and laboratory data were documented. Statistical analysis was done with appropriate clinical tests with significance set at p<0.05. Result(s): Tocilizumab significantly reduced deaths in patients as well as the need for mechanical ventilation with NNT=3 and 5, respectively. The same held true even when the data were adjusted for age, gender, and number of comorbidities. Number of comorbidities had a negative association with mortality and need for mechanical ventilation irrespective of administration of tocilizumab as evidenced by multivariable logistic regression. There was no effect of tocilizumab in shortening the hospital stay in patients. Conclusion(s): Tocilizumab seems to be a promising agent for the treatment of moderate to severe COVID-19 pneumonia and similar agents hold promise for any similar future emerging infections. Copyright © 2022 The Authors.

Implementation of a Physician Incentivized QI Strategy to Address Time to Parenteral Analgesia in Patients with Voe in the Pediatric Emergency Department

Introduction Vaso-occlusive episodes (VOE) are the most common cause of pediatric Emergency Department (ED) visits and hospitalizations in Sickle Cell Disease (SCD). The National Heart Lung and Blood Institute published an Expert Panel Report regarding the management of SCD and VOE. Their consensus statement recommends initiating analgesic therapy within 30 minutes of triage or within 60 minutes of registration in the Emergency Department. Previous studies have demonstrated that earlier maximum opioid has been associated with shorter length of hospitalization and improved time to ED disposition decision. Despite the overwhelming evidence for timely administration of parenteral analgesic, significant delays still exist in delivery of pain medication in the pediatric SCD population. Barriers to timely administration include rapid triage of SCD patients, provider ordering of pain medication, and peripheral intravenous access. Therefore, a standardized approach to pain management may improve ED management of SCD crises. In order to address timely administration of opiates to SCD patients with VOE episodes in our pediatric ED a SCD pain order set was developed. This order set implemented the use of intranasal (IN) fentanyl as a first line analgesic for SCD patients who presents to the ED with VOE. The purpose of this study was addressing barriers to decrease time to parenteral opioid administration in the pediatric ED. Methods This Quality Improvement (QI) measure was performed at a free-standing, urban pediatric ED. Patients were included if they had a diagnosis of SCD and presented with a pain score >5 and without fever. A PDSA cycle was utilized for designing and evaluating the proposed changes. This cycle consisted of three intervention phases: (1) electronic medical record (EMR) order set development in October 2019, (2) provider incentive for order set use in January 2020, and (3) nursing/patient & family education in April 2020. Baseline data was collected pre-intervention from April-September 2019. The outcomes measures were mean time to 1 st analgesic, mean time from triage to disposition, Hospital Length of Stay, and overall admission rates. Our balancing measure included 48 hour ED re-visits after discharge. Results There were 67 ED visits from April-September 2019 (pre-intervention) and 104 ED visits in the post-intervention data from October-June 2020. There was no significant difference in age or initial pain score in the pre- and post- intervention groups. Improvements were seen in: mean time to first analgesic (58 to 26 minutes), time to disposition (271 to 213 minutes). Hospital length of stay was found to increase with the introduction of IN fentanyl: pre-intervention (120 hours), phase 1 (148 hours), phase 2 (152 hours), phase 3 (218 hours). However, the overall admission rate decreased (55% to 44%). The number of 48-hour ED re-visits remained stable. Conclusion By using QI methods to address key barriers in the pediatric ED, we demonstrated that timely administration of parenteral analgesic can be achieved for SCD patients with VOE. Utilizing the EMR order set allowed for more stream-lined care, both by physicians and nursing staff, resulting in more rapid ordering of medication therefore decreasing time to ED disposition. Additional interventions such as provider incentivization to meet the goal of parenteral opioids within 30 minutes of patient arrival led to further improvement. One of the greatest barriers to our QI intervention was hesitancy both by patients and their caregivers regarding the efficacy of IN fentanyl in decreasing pain compared to IV opioid. Further education was needed both for families and medical staff regarding the efficacy of IN fentanyl as a first line analgesic. It is unclear why overall hospital length of stay was not shown to be decreased with these interventions but this can be offset by an overall decrease in hospital admissions seen with our interventions. This data may be limited by the SARS-CoV-2 pandemic and how psychosocial stressors can impact patients with chron c medical conditions. Length of stay is also confounded by other factors during the hospitalization and acquisition of other diagnoses such as acute chest. Future research is needed to determine if the demonstrated trend of admission rates and hospital length of stay can be replicated in other pediatric EDs and whether earlier opioid administration affects the outcome of VOEs beyond the ED. [Formula presented] Disclosures: No relevant conflicts of interest to declare.

COVID-19 and health-care worker's combating approach: an exhausting job demand to satisfy

Purpose: This study aims to investigate the relationship between person-role conflict, psychological capital and emotional exhaustion. Specifically, the research explores how person-role conflict magnified due to daily contact with COVID-19 carriers leads doctors and nurses to experience emotional exhaustion. Moreover, psychological capital function as an explanatory mechanism between stressor strain relationships has also been tested. Design/methodology/approach: The study results are based on three months of lagged data conducted from the sample of 347 frontline physicians and nurses who provide treatment and care to infected people. To test direct, indirect and total effect, the author's used PROCESS Macro. Findings: The results suggested that person-role conflict reduces state-like psychological capital and increases emotional exhaustion through reduced psychological capital. Results aligned with the model's expectations in that psychological capital mediated the relationship between person-role conflict and emotional exhaustion, and the mediation was partial. Originality/value: This paper is the first one that tested the link between person-role conflict and emotional exhaustion. Moreover, up till now, no study has examined the mediating role of psychological capital in the relationship between person-role conflict and emotional exhaustion. Finally, in the context of the contagion outbreak, this is the preliminary effort that validated the resource loss cycle principle of conservation of resource theory. © 2021, Emerald Publishing Limited.