Lightweight Method for the Rapid Diagnosis of Coronavirus Disease 2019 from Chest X-ray Images using Deep Learning Technique

A rapid screening method is required for screening coronavirus disease 2019 (COVID-19) patients. Therefore, we proposed a model based on DenseNet-201 to detect and differentiate COVID-19 patients from normal people and patients with other bacterial/viral cases of pneumonia using chest X-ray images. Our four-class model was found to have an accuracy of 91.01 ± 1.86 (mean ± standard deviation) and a sensitivity of 92.65 ± 1.28 using a five-fold cross-validation method. Moreover, it was a relatively lightweight and robust model with a simplified structure and fewer parameters, training, and testing epochs. As a supplementary diagnosis tool, physicians can detect COVID-19 faster using this model. © 2021 IEEE.

Covidense: Providing a suitable solution for diagnosing Covid-19 lung infection based on deep learning from chest X-ray images of patients

Purpose: Coronavirus disease 2019 (Covid-19), first reported in December 2019 in Wuhan, China, has become a pandemic. Chest imaging is used for the diagnosis of Covid-19 patients and can address problems concerning Reverse Transcription-Polymerase Chain Reaction (RT-PCR) shortcomings. Chest X-ray images can act as an appropriate alternative to Computed Tomography (CT) for diagnosing Covid-19. The purpose of this study is to use a Deep Learning method for diagnosing Covid-19 cases using chest X-ray images. Thus, we propose Covidense based on the pre-trained Densenet-201 model and is trained on a dataset comprising chest X-ray images of Covid-19, normal, bacterial pneumonia, and viral pneumonia cases. Materials and Methods: In this study, a total number of 1280 chest X-ray images of Covid-19, normal, bacterial and viral pneumonia cases were collected from open access repositories. Covidense, a convolutional neural network model, is based on the pre-trained DenseNet-201 architecture, and after pre-processing the images, it has been trained and tested on the images using the 5-fold cross-validation method. Results: The accuracy of different classifications including classification of two classes (Covid-19, normal), three classes 1 (Covid-19, normal and bacterial pneumonia), three classes 2 (Covid-19, normal and viral pneumonia), and four classes (Covid-19, normal, bacterial pneumonia and viral pneumonia) are 99.46%, 92.86%, 93.91 %, and 91.01% respectively. Conclusion: This model can differentiate pneumonia caused by Covid-19 from other types of pneumonia, including bacterial and viral. The proposed model offers high accuracy and can be of great help for effective screening. Thus, reducing the rate of infection spread. Also, it can act as a complementary tool for the detection and diagnosis of Covid-19. Copyright © 2021 Tehran University of Medical Sciences.

The clinical course and laboratory results changes in icu-admitted patients with covid-19 since the beginning of hospitalization to icu admission

In this study, we aimed to evaluate the changes in laboratory parameters of COVID-19 hospitalized patients who admitted to the intensive care unit (ICU). In this retrospective study, the confirmed cases of COVID-19 patients who were hospitalized in ward and ICU from 19 January 2020 to 27 February 2020 in Firoozgar hospital, Tehran, Iran were enrolled. We analyzed clinical characteristics and laboratory findings through medical records. SPSS v.25 was used for statistical analysis. The 70 COVID-19 patients by the mean age±std. deviation 68.37±13.29 years (range 27-93 years) were carried out. The average duration of hospitalization±std. deviation was 7.4±6.17 days (7-27 days). 43 cases were male (61.4%). Fifteen patients (21.4%) did not have any underlying disease. There was significant increasing in laboratory parameters included white blood cells (p<0.0001), Creatinine (p = 0.007), aspartate aminotransferase (AST) (p = 0.050) and alanine transaminase (ALT) (p = 0.031). However, lymphocyte count was significantly decreased during hospitalization in ward before ICU-admission (p<0.0001). There was no significant difference for platelets count (p = 0.94), lactate dehydrogenase (p = 0.36), International Normalized Ratio (INR) (p = 0.114) and Partial Thromboplastin Time (PTT) (p = 0.72). Most ICU-admitted patients presented with respiratory syndrome characteristics. ICU-admitted patients had significant increase in WBC and decrease in lymphocyte count. Evidence of failure in kidney, liver function, higher activity of the coagulation system were discovered among ICU-admitted patients.

Liver enzymes and inpatient deaths in COVID-19 hospitalized patients

Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [COVID-19] quickly turned into a pandemic. Gastrointestinal involvement, especially liver diseases, is one of the main complications of COVID-19 patients. Objectives: The current study aimed to evaluate the high incidence of liver involvement in COVID-19 hospitalized patients and its association with mortality. Methods: A total of 560 hospitalized patients with a confirmed diagnosis of COVID-19 were included. Death was considered as the outcome. In addition to liver enzymes, demographic, clinical, and other laboratory data were also collected. Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels_ 40 were considered as abnormal. To investigate the association between abnormal levels of liver enzymes and death, multiple regression logistic was used. Results: According to the findings, 29.1% (95% CI = 25.3% - 32.9%) of patients had high levels (_ 40 IU) of ALT, and 45.1% (95% CI = 40.9% - 49.3%) had high levels of AST (_ 40 IU). The frequency (based on %) of high levels of AST (_ 40 U/liter) was significantly higher in patients who died [67.3% (95% CI = 54.5% - 80.1%] of COVID-19 than those who survived [44.9% (95% CI = 39.7% - 50.0%)] (Pvalue < 0.001). No significant difference was detected in ALT between expired [29.1% (95% CI = 16.7% - 41.5%)] and survived patients [30.7% (95% CI = 25.9% - 35.5%] (P-value = 0.791). AST was found to have an independent association with death in multiple logistic regression (Wald = 4.429, OR (95% CI) = 1.014 (1.008 - 1.020), P-value = 0.035). Conclusions: Liver involvement is a common finding in COVID-19 hospitalized patients. Higher levels of AST were significantly associated with an increased mortality rate in COVID-19 patients.

Opiate Use in Hospitalized Patients with IBD Can Be Significantly Reduced Using a Proactive Pain Protocol: Results of a Randomized Controlled Trial

INTRODUCTION: Hospitalized patients with inflammatory bowel disease (IBD) are often treated with high doses of opioids, which can lead to opioid dependence, decreased quality of life, and increased mortality. We developed an evidence-based inpatient pain protocol for adults with inflammatory bowel disease (IBD) comprised of scheduled acetaminophen, celecoxib, gabapentin, and as-needed lorazepam (Table 1). In this study, we compared this proactive pain protocol to usual care in a randomized control trial. METHODS: Hospitalized, nonpregnant adults with IBD with abdominal pain and without recent surgery were randomized to the proactive pain protocol or to a standard-of-care reactive pain regimen (as-needed acetaminophen and opioids). Outcomes included daily pain (assessed by numeric rating scores, 0-10), average daily morphine milligram equivalents (MME), length of stay (LOS), need for surgery during admission, and 30-day readmission rates. Intended sample size was 166 subjects, but enrollment was halted early due to lower than expected recruitment and COVID-19 research restrictions. Subjects were analyzed per-protocol. RESULTS: Thirty-three subjects were enrolled;1 withdrew consent and was excluded from analysis. Seventeen were randomized to the proactive protocol and 15 to the reactive regimen (control group). One subject in the control group received the proactive protocol. Baseline demographics, race, type of IBD, CRP, and reason for admission were similar between the two groups. There was a significant decrease in pain over time in both groups (22.8 ± 2.8 points, P < 0.0001). Overall, those receiving the proactive protocol had numerically lower pain scores over the course of hospitalization (3.02 ± 0.90 vs 4.29 ± 0.81, P = 0.059) (Figure 1) and consumed fewer daily MME than controls (13.94 ± 5.96 vs 37.26 ± 10.51, P = 0.02) (Figure 2). There were no differences in LOS (7.3 ± 6.6 vs 7.1 ± 3.5, P = 0.66), surgery during admission (11.1% vs 21.4%, P = 0.63), and readmission (11.1% vs 14.3%, P . 0.99) between the two groups. One subject had emesis after taking celecoxib which stopped after discontinuation;no other adverse events were noted. CONCLUSION: A proactive pain protocol reduces the use of opioids and may also improve overall pain control compared with a standard, reactive pain regimen in hospitalized patients with IBD. Proactive pain control with scheduled non-opioid pain medications should be considered for patients hospitalized with IBD to reduce reliance on opioids.