Between January and October of 2020, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus has infected more than 34 million persons in a worldwide pandemic leading to over one million deaths worldwide (data from the Johns Hopkins University). Since the virus begun to spread, emergency departments were busy with COVID-19 patients for whom a quick decision regarding in-or outpatient care was required. The virus can cause characteristic abnormalities in chest radiographs (CXR), but, due to the low sensitivity of CXR, additional variables and criteria are needed to accurately predict risk. Here, we describe a computerized system primarily aimed at extracting the most relevant radiological, clinical, and laboratory variables for improving patient risk prediction, and secondarily at presenting an explainable machine learning system, which may provide simple decision criteria to be used by clinicians as a support for assessing patient risk. To achieve robust and reliable variable selection, Boruta and Random Forest (RF) are combined in a 10-fold cross-validation scheme to produce a variable importance estimate not biased by the presence of surrogates. The most important variables are then selected to train a RF classifier, whose rules may be extracted, simplified, and pruned to finally build an associative tree, particularly appealing for its simplicity. Results show that the radiological score automatically computed through a neural network is highly correlated with the score computed by radiologists, and that laboratory variables, together with the number of comorbidities, aid risk prediction. The prediction performance of our approach was compared to that that of generalized linear models and shown to be effective and robust. The proposed machine learning-based computational system can be easily deployed and used in emergency departments for rapid and accurate risk prediction in COVID-19 patients.
Background In March 2020 we faced a huge spread of the epidemic of SARS-CoV2 in northern Italy; the Emergency Departments (ED) and the Emergency Medical Services (EMS) were overwhelmed by patients requiring care. The hospitals were forced to reorganize their services, and the ED was the focal point of this challenge. As Emergency Department in a metropolitan area of the region most affected, we saw an increasing number of patients with COVID-19, and we made some structural and staff implementations according to the evolution of the epidemic. Methods We analysed in a narrative way the weaknesses and the point of strength of our response to COVID-19 first outbreak, focusing point by point on main challenges and minor details involved in our ED response to the pandemics. Results The main stems for our response to the pandemic were: use of clear and shared contingency plans, as long as preparedness to implement them; stockage of as much as useful material can be stocked; training of the personnel to be prepared for a fast response, trying to maintain divided pathway for COVID-19 and non-COVID-19 patients, well-done isolation is a key factor; preparedness to de-escalate as soon as needed. Conclusions We evaluated our experience and analysed the weakness and strength of our first response to share it with the rest of the scientific community and colleagues worldwide, hoping to facilitate others who will face the same challenge or similar challenges in the future. Shared experience is the best way to learn and to avoid making the same mistakes.
Background The scientific evidence regarding the risk of delayed intracranial bleeding (DB) after mild traumatic brain injury (MTBI) in patients administered an antiplatelet agent (APA) is scant and incomplete. In addition, no consensus exists on the utility of a routine repeated head computed tomography (CT) scan in these patients. Objective The aim of this study was to evaluate the risk of DB after MTBI in patients administered an APA. Methods A systematic review and meta-analysis of prospective and retrospective observational studies enrolling adult patients with MTBI administered an APA and who had a second CT scan performed or a clinical follow-up to detect any DB after a first negative head CT scan were conducted. The primary outcome was the risk of DB in MTBI patients administered an APA. The secondary outcome was the risk of clinically relevant DB (defined as any DB leading to neurosurgical intervention or death). Results Sixteen studies comprising 2930 patients were included in this meta-analysis. The pooled absolute risk for DB was 0.77% (95% CI 0.23–1.52%), ranging from 0 to 4%, with substantial heterogeneity (I2 = 61%). The pooled incidence of clinically relevant DB was 0.18%. The subgroup of patients on dual antiplatelet therapy (DAPT) had an increased DB risk, compared to the acetylsalicylic acid (ASA)-only patients (2.64% vs. 0.22%; p = 0.04). Conclusion Our systematic review showed a very low risk of DB in MTBI patients on antiplatelet therapy. We believe that such a low rate of DB could not justify routine repeated CT scans in MTBI patients administered a single APA. We speculate that in the case of clinically stable patients, a repeated head CT scan could be useful for select high-risk patients and for patients on DAPT before discharge.
Purpose: To determine the performance of a chest radiograph (CXR) severity scoring system combined with clinical and laboratory data in predicting the outcome of COVID-19 patients. Materials and Methods: We retrospectively enrolled 301 patients who had reverse transcriptase-polymerase chain reaction (RT-PCR) positive results for COVID-19. CXRs, clinical and laboratory data were collected. A CXR severity scoring system based on a qualitative evaluation by two expert thoracic radiologists was defined. Based on the clinical outcome, the patients were divided into two classes: moderate/mild (patients who did not die or were not intubated) and severe (patients who were intubated and/or died). ROC curve analysis was applied to identify the cutoff point maximizing the Youden index in the prediction of the outcome. Clinical and laboratory data were analyzed through Boruta and Random Forest classifiers. Results: The agreement between the two radiologist scores was substantial (kappa = 0.76).
Background Bleeding is an important cause of death in trauma victims. In 2010, the CRASH-2 study, a multicentre randomized control trial on the effect of tranexamic acid (TXA) administration to trauma patients with suspected significant bleeding, reported a decreased mortality in randomized patients compared to placebo. Currently, no evidence on the use of TXA in humanitarian, low-resource settings is available. We aimed to measure the hospital outcomes of adult patients with severe traumatic bleeding in the Médecins Sans Frontières Tabarre Trauma Centre in Port-au-Prince, Haiti, before and after the implementation of a Massive Haemorrhage protocol including systematic early administration of TXA. Methods Patients admitted over comparable periods of four months (December2015- March2016 and December2016 - March2017) before and after the implementation of the Massive Haemorrhage protocol were investigated. Included patients had blunt or penetrating trauma, a South Africa Triage Score ≥ 7, were aged 18–65 years and were admitted within 3 h from the traumatic event. Measured outcomes were hospital mortality and early mortality rates, in-hospital time to discharge and time to discharge from intensive care unit. Results One-hundred and sixteen patients met inclusion criteria. Patients treated after the introduction of the Massive Haemorrhage protocol had about 70% less chance of death during hospitalization compared to the group “before” (adjusted odds ratio 0.3, 95%confidence interval 0.1–0.8). They also had a significantly shorter hospital length of stay (p = 0.02). Conclusions Implementing a Massive Haemorrhage protocol including early administration of TXA was associated with the reduced mortality and hospital stay of severe adult blunt and penetrating trauma patients in a context with poor resources and limited availability of blood products.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.