Search

Explainable artificial intelligence (AI) focuses on developing models and algorithms that provide transparent and interpretable insights into decision-making processes. By elucidating the reasoning behind AI-driven diagnoses and treatment recommendations, explainability can gain the trust of healthcare experts and assist them in difficult diagnostic tasks. Sepsis is characterized as a serious condition that happens when the immune system of the body has an extreme response to an infection, causing tissue and organ damage and leading to death. Physicians face challenges in diagnosing and treating sepsis due to its complex pathogenesis. This work aims to provide an overview of the recent studies that propose explainable AI models in the prediction of sepsis onset and sepsis mortality using intensive care data. The general findings showed that explainable AI can provide the most significant features guiding the decision-making process of the model. Future research will investigate explainability through argumentation theory using intensive care data focused on sepsis patients.

Introduction The presence of a central venous catheter (CVC) leads to a high risk for blood infections, which are associated with increases in morbidity, mortality, and costs. This study aims to assess intensive care unit (ICU) nurses' and physicians' knowledge regarding the Centers for Disease Control and Prevention (CDC) guidelines for preventing CVC-related infections before and after an interactive distance education delivered through the e-learning platform Teleprometheus. Materials and methods The study was conducted among 85 nurses and physicians in Nicosia's General Hospital Intensive Care Unit (NGH-ICU) and high dependency unit (HDU). A validated questionnaire was used to assess nurses' and physicians' knowledge. Results Prior to the online interactive distance education, the mean total knowledge score was x̄ = 4.8 (SD = 2.46), while after, the mean total knowledge score increased to x̄ = 8.9 (SD = 2.38) (p<0.001). ICU physicians had a higher mean total knowledge score (x̄ = 10.20) than ICU nurses (x̄ = 8.75) after the intervention. There was no correlation between years of experience in the ICU and the level of knowledge (r = 0.048). The interactive distance education was positively evaluated by the participants, through a questionnaire, specially designed for this study. Discussion The most important findings were that (a) the level of knowledge of the participants improved with a statistically significant difference after the completion of the e-course, (b) the level of knowledge of the participants, after the completion of the e-course, was much higher from other studies, (c) there was no correlation between the years of experience of ICU health professionals and their level of knowledge, and (d) the interactive distance e-course was positively evaluated and satisfied the participants. Conclusion The current study demonstrates that in high-intensity work environments, such as ICUs, adopting e-learning approaches seems more necessary than ever., Competing Interests: The authors have declared that no competing interests exist., (Copyright © 2023, Foka et al.)

In any infectious disease, understanding the modes of transmission is key to selecting effective public health measures. In the case of COVID-19 spread, the strictness of the imposed measures outlined the lack of understanding on how SARS-CoV-2 transmits, particularly via airborne pathways. With the aim to characterize the transmission dynamics of airborne SARS-CoV-2, 165 and 62 air and environmental samples, respectively, were collected in four COVID-19 wards and ICUs in Cyprus and analyzed by RT-PCR. An alternative method for SARS-CoV-2 detection in air that provides comparable results but is less cumbersome and time demanding, is also proposed. Considering that all clinics employed 14 regenerations per hour of full fresh air inside patient rooms, it was hypothesized that the viral levels and the frequency of positive samples would be minimum outside of the rooms. However, it is shown that leaving the door opened in patient rooms hinders the efficiency of the ventilation system applied, allowing the virus to escape. As a result, the highest observed viral levels (135 copies m -3 ) were observed in the corridor of a ward and the frequency of positive samples in the same area was comparable to that inside a two-bed cohort. SARS-CoV-2 in that corridor was found primarily to lie in the coarse mode, at sizes between 1.8 and 10 μm. Similar to previous studies, the frequency of positive samples and viral levels were the lowest inside intensive care units. However, if a patient with sufficient viral load (Ct-value 31) underwent aerosol generating procedures, positive samples with viral levels below 45 copies m -3 were acquired within a 2 m distance of the patient. Our results suggest that a robust ventilation system can prevent unnecessary exposure to SARS-CoV-2 but with limitations related to foot traffic or the operations taking place at the time., Competing Interests: The authors declare the following conflict of interests: Lakis Palazis and Maria Foka work at the ICU of Nicosia General Hospital. All other authors declare no competing interests., (© 2023 The Authors.)

Background: Device-associated health care-associated infections (DA-HAIs) are a major threat to patient safety, particularly in the Intensive Care Unit (ICU). This study aimed to evaluate the effectiveness of a bundle of infection control measures to reduce DA-HAIs in the ICU of a General Hospital in the Republic of Cyprus, over a 3-year period., Methods: We studied 599 ICU patients with a length of stay (LOS) for at least 48 hours. Our prospective cohort study was divided into three surveillance phases. Ventilator-associated pneumonia (VAP), central line-associated blood-stream infections (CLABSI), and catheter-associated blood-stream infections (CAUTI) incidence rates, LOS, and mortality were calculated before, during, and after the infection prevention and control programme., Results: There was a statistically significant reduction in the number of DA-HAI events during the surveillance periods, associated with DA-HAIs prevention efforts. In 2015 (prior to programme implementation), the baseline DA-HAIs instances were 43: 16 VAP (10.1/1000 Device Days), 21 (15.9/1000DD) CLABSIs, and 6 (2.66/1000DD) CAUTIs, (n = 198). During the second phase (2016), CLABSIs prevention measures were implemented and the number of infections were 24: 14 VAP (12.21/1000DD), 4 (4.2/1000DD) CLABSIs, and 6 (3.22/1000DD) CAUTIs, (n = 184). During the third phase (2017), VAP and CAUTI prevention measures were again implemented and the rates were 6: (3 VAP: 12.21/1000DD), 2 (1.95/1000DD) CLABSIs, and 1 (0.41/1000DD) CAUTIs, (n = 217). There was an overall reduction of 87% in the total number of DA-HAIs instances for the period 1 January 2015 to 31 December 2017., Conclusions: The significant overall reduction in DA-HAI rates indicates that a comprehensive infection control programme can affect DA-HAI rates., (© 2020 British Association of Critical Care Nurses.)