Robust genetic machine learning ensemble model for intrusion detection in network traffic.

Network security has developed as a critical research subject as a result of the Rapid advancements in the development of Internet and communication technologies over the previous decades. The expansion of networks and data has caused cyber-attacks on the systems, making it difficult for network security to detect breaches effectively. Current Intrusion Detection Systems (IDS) have several flaws, including their inability to prevent attacks on their own, the requirement for a professional engineer to administer them, and the occurrence of false alerts. As a result, a plethora of new attacks are being created, making it harder for network security to properly detect breaches. Despite the best efforts, IDS continues to struggle with increasing detection accuracy while lowering false alarm rates and detecting new intrusions. Therefore, network intrusion detection enhancement by preprocessing and generation of highly reliable algorithms is the main focus nowadays. Machine learning (ML) based IDS systems have recently been implemented as viable solutions for quickly detecting intrusions across the network. In this study, we use a combined data analysis technique with four Robust Machine learning ensemble algorithms, including the Voting Classifier, Bagging Classifier, Gradient Boosting Classifier, and Random Forest-based Bagging algorithm along with the proposed Robust genetic ensemble classifier. For each algorithm, a model is created and tested using a Network Dataset. To assess the performance of both algorithms in terms of their ability to anticipate the anomaly occurrence, graphs of performance rates have been evaluated. The suggested algorithm outperformed other methods as it shows the lowest values of mean square error (MSE) and mean absolute error (MAE). The experiments were conducted on the Network traffic dataset available on Kaggle, on the Python platform, which has limited samples. The proposed method can be applied in the future with more machine learning ensemble classifiers and deep learning techniques.

Characteristics and outcome of tertiary care critically ill COVID-19 patients with multiple comorbidities admitted to the intensive care unit.

PURPOSE: We conducted this study to evaluate the characteristics and outcomes exclusively in high-risk coronavirus disease 2019 (COVID-19) tertiary care patients with multiple comorbidities, as very few have reported outcomes in this specific cohort. METHODS: All patients, with two or more risk factors for COVID-19 and Charlson Comorbidity Index (CCI) of >2, who were admitted to intensive care unit (ICU) between March and December 2020 were included. Their characteristics, ICU course, and outcomes as well as differences between nonsurvivors and survivors were evaluated. The primary outcome was all-cause 28-day mortality. RESULTS: Out of 1152 COVID-19 patients, 101 met the inclusion criteria. The patients had an average of 4 or more comorbidities with a very high CCI of 5. The 28-day all-cause mortality was 23% and inhospital mortality was 32%. Among all risk factors, only age > 70 years, male gender, and chronic kidney disease were significant determinants of mortality (P < 0.03). Admission PaO2/FiO2 ratio and elevated inflammatory markers were same among survivors and nonsurvivors (P > 0.66). The mean time from presentation to ICU admission (59 vs. 38 h), APACHE II score (20.5 vs. 17), ICU length of stay (25 vs. 12 days), and hospital length of stay (28 vs. 20 days) were all higher in nonsurvivors as compared to survivors, respectively (P < 0.03). Fifty-four percent of the patients were intubated and had higher 28-day (40%) and inhospital (55%) mortality. CONCLUSION: Tertiary care patients with multiple comorbidities have higher mortality than what is reported for mixed populations. Further studies are needed to determine realistic mortality benchmarks for these patients.

Too early for admission? A Telemedicine follow up comparison of mild COVID-19 patients from the Emergency Department and Clinics.

OBJECTIVE: Mildly symptomatic COVID-19 patients may seek medical attention either in the Emergency Department (ED) or Ambulatory Clinics (AC). However, it is unclear if ED patients have different characteristics and outcomes than AC patients when discharged under telemedicine surveillance, which we explored in this study. METHODS: Patients with mild or asymptomatic COVID-19 disease referred to a multidisciplinary Telemedicine clinical service (TM-CS) program in an urban tertiary-care hospital, between June 2020 and February 2021, were evaluated. Those referred from ED were labeled "ED Group" and ones from AC as "AC Group." Their characteristics, clinical features and outcomes including telemedicine parameters, subsequent ED visits, hospital admission, oxygen requirements, intensive care unit (ICU) admission, and mortality were compared. RESULTS: Out of 1132 confirmed non-admitted COVID-19 patients, 526 with mild (89%) or asymptomatic (11%) disease were enrolled in TM-CS. Majority of these were referred from ED (n = 370; 70%) and rest (n = 156, 30%) from the AC. Patients in the ED group compared to AC group, had higher BMI (28.9 vs. 27.5), higher Charlson Comorbidity Index (1.4 vs. 0.9), and higher incidence of comorbidities (50% vs. 22%), P ≤ 0.01. However, there were no differences in the ED and AC groups in subsequent ED visits (26% vs. 24%), hospital admission (18% vs. 15%), oxygen requirements (5% vs. 4%), ICU admission (1% vs. 2%), and mortality (0.3% vs. 0.6%), respectively (P > 0.40). CONCLUSION: Significant number of mild COVID-19 patients head to the ED for initial assistance but have similar outcomes to AC patients. TM-CS could be a safe alternative for follow-up monitoring of these patients.

From Epidemic to Pandemic: Comparing Hospital Staff Emotional Experience Between MERS and COVID-19.

Objective: Both Middle East Respiratory Syndrome (MERS) and Coronavirus Disease 2019 (COVID-19) have an emotional toll on healthcare workers (HCWs), but the difference of the impact between the two diseases remains unknown.Design: A cross sectional descriptive survey.Setting: A tertiary care hospital.Participants: 125 HCWs who worked during the 2014 MERS as well as the 2020 COVID-19 outbreaks in high-risk areas of the hospital including critical care, emergency room and COVID-19 clinics.Methods: The comprehensive survey comprised 5 sections and 68 questions and was administered to HCWs before availability of the COVID-19 vaccine. The survey evaluated hospital staff emotions, perceived stressors, external factors that reduced stress, personal coping strategies, and motivators for future outbreaks. The participants rated each question for MERS and COVID-19 simultaneously on a scale from 0-3. The responses were reported as mean and standard deviation, while Wilcoxon signed-rank test was used to calculate the difference in responses.Results: There were 102 (82%) participants who returned the questionnaire. The ritual of obsessive hand washing, emotional and physical fatigue, ongoing changes in infection control guidelines, fear of community transmission, and limitations on socialization and travel were the major stressors that were significantly worse during COVID-19 compared to MERS (P<0.05) and led to HCWs adoption of additional 'personal' coping strategies during COVID-19. There was no difference between COVID-19 and MERS, however, among preferences for 'external' factors made available to HCWs that could reduce stress or in their preferences for motivators to work in future outbreaks (P>.05).Conclusion: Both the MERS and COVID-19 outbreaks were emotionally draining for HCWs. However, COVID-19 was a relatively more stressful experience than MERS for HCWs and led to greater personal, behavioral, and protective adaptations by the hospital staff.

Comparison of characteristics and ventilatory course between coronavirus disease 2019 and Middle East respiratory syndrome patients with acute respiratory distress syndrome.

BACKGROUND: Both coronavirus disease 2019 (COVID-19) and Middle East respiratory syndrome (MERS) can cause acute respiratory distress syndrome (ARDS); however, their ARDS course and characteristics have not been compared, which we evaluate in our study. METHODS: MERS patients with ARDS seen during the 2014 outbreak and COVID-19 patients with ARDS admitted between March and December 2020 in our hospital were included, and their clinical characteristics, ventilatory course, and outcomes were compared. RESULTS: Forty-nine and 14 patients met the inclusion criteria for ARDS in the COVID-19 and MERS groups, respectively. Both groups had a median of four comorbidities with high Charlson comorbidity index value of 5 points (P>0.22). COVID-19 patients were older, obese, had significantly higher initial C-reactive protein (CRP), more likely to get trial of high-flow oxygen, and had delayed intubation (P≤0.04). The postintubation course was similar between the groups. Patients in both groups experienced a prolonged duration of mechanical ventilation, and majority received paralytics, dialysis, and vasopressor agents (P>0.28). The respiratory and ventilatory parameters after intubation (including tidal volume, fraction of inspired oxygen, peak and plateau pressures) and their progression over 3 weeks were similar (P>0.05). Rates of mortality in the ICU (53% vs. 64%) and hospital (59% vs. 64%) among COVID-19 and MERS patients (P≥0.54) were very high. CONCLUSIONS: Despite some distinctive differences between COVID-19 and MERS patients prior to intubation, the respiratory and ventilatory parameters postintubation were not different. The higher initial CRP level in COVID-19 patients may explain the steroid responsiveness in this population.

Comparison of Clinical Characteristics and End-of-Life Care Between COVID-19 and Non-COVID-19 Muslim Patients During the 2020 Pandemic.

BACKGROUND: Little is known about end-of-life care among Muslim patients, particularly during Coronavirus disease 2019 (COVID) pandemic, which we report here. METHODS: The clinical characteristics, end-of-life care and resuscitation status of Muslim patients who died in the ICU of our tertiary care hospital in year 2020 from COVID were compared to Non-COVID patients. RESULTS: There were 32 patients in COVID and 64 in the Non-COVID group. A major proportion, mainly of Non-COVID patients, already had a hospice eligible terminal disease at baseline (p=.002). COVID patients were admitted to the ICU sooner after hospitalization (2.2 vs. 17 days), had prolonged duration of mechanical ventilation (18.5 vs. 6 days) and longer ICU stay (24 vs. 8 days) than non-COVID patients, respectively (p<.001). Almost all patients were "Full Code" initially. However, status was eventually changed to 'do-not-attempt resuscitation' (DNAR) in about 60% of the cohort. COVID patients were made DNAR late in their ICU stay, predominantly in the last 24 hours of life (p=.04). Until the very end, patients in both groups were on tube feeds, underwent blood draws and imaging, required high dose vasopressors, with few limitations or withdrawal of therapies. Family members were usually not present at bedside at time of death. There was minimal involvement of chaplain and palliative care services. CONCLUSIONS: Muslim COVID-19 patients had prolonged mechanical ventilation and ICU stay and a delayed decision to DNAR status than non-COVID Muslim patients. Limitation or withdrawal of therapy occurred infrequently. The utilization of chaplain and palliative care service needs improvement.