Improved patient mortality predictions in emergency departments with deep learning data-synthesis and ensemble models.

The triage process in emergency departments (EDs) relies on the subjective assessment of medical practitioners, making it unreliable in certain aspects. There is a need for a more accurate and objective algorithm to determine the urgency of patients. This paper explores the application of advanced data-synthesis algorithms, machine learning (ML) algorithms, and ensemble models to predict patient mortality. Patients predicted to be at risk of mortality are in a highly critical condition, signifying an urgent need for immediate medical intervention. This paper aims to determine the most effective method for predicting mortality by enhancing the F1 score while maintaining high area under the receiver operating characteristic curve (AUC) score. This study used a dataset of 7325 patients who visited the Yonsei Severance Hospital's ED, located in Seoul, South Korea. The patients were divided into two groups: patients who deceased in the ED and patients who didn't. Various data-synthesis techniques, such as SMOTE, ADASYN, CTGAN, TVAE, CopulaGAN, and Gaussian Copula, were deployed to generate synthetic patient data. Twenty two ML models were then utilized, including tree-based algorithms like Decision tree, AdaBoost, LightGBM, CatBoost, XGBoost, NGBoost, TabNet, which are deep neural network algorithms, and statistical algorithms such as Support Vector Machine, Logistic Regression, Random Forest, k-nearest neighbors, and Gaussian Naive Bayes, as well as Ensemble Models which use the results from the ML models. Based on 21 patient information features used in the pandemic influenza triage algorithm (PITA), the models explained previously were applied to aim for the prediction of patient mortality. In evaluating ML algorithms using an imbalanced medical dataset, conventional metrics like accuracy scores or AUC can be misleading. This paper emphasizes the importance of using the F1 score as the primary performance measure, focusing on recall and specificity in detecting patient mortality. The highest-ranked model for predicting mortality utilized the Gaussian Copula data-synthesis technique and the CatBoost classifier, achieving an AUC of 0.9731 and an F1 score of 0.7059. These findings highlight the effectiveness of machine learning algorithms and data-synthesis techniques in improving the prediction performance of mortality in EDs.

Efficacy of a four-tier infection response system in the emergency department during the coronavirus disease-2019 outbreak.

INTRODUCTION: The coronavirus disease (COVID-19) pandemic has delayed the management of other serious medical conditions. This study presents an efficient method to prevent the degradation of the quality of diagnosis and treatment of other critical diseases during the pandemic. METHODS: We performed a retrospective observational study. The primary outcome was ED length of stay (ED LOS). The secondary outcomes were the door-to-balloon time in patients with suspected ST-segment elevation myocardial infarction and door-to-brain computed tomography time for patients with suspected stroke. The outcome measures were compared between patients who were treated in the red and orange zones designated as the changeable isolation unit and those who were treated in the non-isolation care unit. To control confounding factors, we performed propensity score matching, following which, outcomes were analyzed for non-inferiority. RESULTS: The mean ED LOS for hospitalized patients in the isolation and non-isolation care units were 406.5 min (standard deviation [SD], 237.9) and 360.2 min (SD, 226.4), respectively. The mean difference between the groups indicated non-inferiority of the isolation care unit (p = 0.037) but not in the patients discharged from the ED (p>0.999). The mean difference in the ED LOS for patients admitted to the ICU between the isolation and non-isolation care units was -22.0 min (p = 0.009). The mean difference in the door-to-brain computed tomography time between patients with suspected stroke in the isolation and non-isolation care units was 7.4 min for those with confirmed stroke (p = 0.013), and -20.1 min for those who were discharged (p = 0.012). The mean difference in the door-to-balloon time between patients who underwent coronary angiography in the isolation and non-isolation care units was -2.1 min (p<0.001). CONCLUSIONS: Appropriate and efficient handling of a properly planned ED plays a key role in improving the quality of medical care for other critical diseases during the COVID-19 outbreak.

Development and evaluation of a multimodality simulation disaster education and training program for hospital nurses.

AIM: To develop a multimodality simulation program for hospital nurses to enhance their disaster competency and evaluate the effect of the program. METHODS: The program implementation started in October 2016 and ended in December 2016. It was developed using the ADDIE model (analysis, design, development, implementation, and evaluation). Evaluation consisted of formative assessment and summative assessment. Formative assessment was performed during triage, crisis management, and problem-solving simulation programs through direct feedback and debriefing from the teacher. Summative assessment was performed using the Kirkpatrick curriculum evaluation framework. RESULTS: Needs assessment using the modified Delphi survey resulted in these competencies for hospital disaster nursing: triage, incident command, surge capacity, life-saving procedures, and special situations. Each competency was matched with the appropriate simulation modalities. A total of 40 emergency nurses participated in the study program. The evaluation of the program resulted in improvement in perception, crisis management, problem solving, and technical skills in disaster nursing. CONCLUSION: Multimodality simulation training program was developed to enhance the competency of hospital nurses in disaster response. All participants improved their disaster response competencies significantly. The program that was developed in this study could be used as a fundamental tool in future research in disaster curriculum development.

International Nursing: Needs Assessment for Training in Disaster Preparedness for Hospital Nurses: A Modified Delphi Study.

Despite high awareness of the need, opportunities for nurses to gain disaster experience or training are limited. In Korea, most disaster training is done in an undergraduate curriculum where there is very limited practice, and the educational topics are mostly focused on the field aspect of disaster events. The purpose of this study was to determine the need for such training for hospital nurses and to determine appropriate and relevant components of the training contents. A qualitative survey approach using the modified Delphi method was used to collect and analyze the data. The surveys were conducted in 3 rounds. After the results were analyzed from the third-round survey, the authors finalized the contents for a training program to prepare nurses for their roles during disasters. Through a structured needs analysis using a modified Delphi survey, the framework for the content development of disaster training curriculum for hospital nurses was developed.

The intrinsic defect structure of exfoliated MoS2 single layers revealed by Scanning Tunneling Microscopy.

MoS2 single layers have recently emerged as strong competitors of graphene in electronic and optoelectronic device applications due to their intrinsic direct bandgap. However, transport measurements reveal the crucial role of defect-induced electronic states, pointing out the fundamental importance of characterizing their intrinsic defect structure. Transmission Electron Microscopy (TEM) is able to image atomic scale defects in MoS2 single layers, but the imaged defect structure is far from the one probed in the electronic devices, as the defect density and distribution are substantially altered during the TEM imaging. Here, we report that under special imaging conditions, STM measurements can fully resolve the native atomic scale defect structure of MoS2 single layers. Our STM investigations clearly resolve a high intrinsic concentration of individual sulfur atom vacancies, and experimentally identify the nature of the defect induced electronic mid-gap states, by combining topographic STM images with ab intio calculations. Experimental data on the intrinsic defect structure and the associated defect-bound electronic states that can be directly used for the interpretation of transport measurements are essential to fully understand the operation, reliability and performance limitations of realistic electronic devices based on MoS2 single layers.

Relief of Injection Pain During Delivery of Local Anesthesia by Computer-Controlled Anesthetic Delivery System for Periodontal Surgery: Randomized Clinical Controlled Trial.

BACKGROUND: Pain from local anesthetic injection makes patients anxious when visiting a dental clinic. This study aims to determine differences in pain according to types of local anesthetizing methods and to identify the possible contributing factors (e.g., dental anxiety, stress, and sex). METHODS: Thirty-one patients who underwent open-flap debridement in maxillary premolar and molar areas during treatment for chronic periodontitis were evaluated for this study. A randomized, split-mouth, single-masked clinical trial was implemented. The dental anxiety scale (DAS) and perceived stress scale (PSS) were administered before surgery. Two lidocaine ampules for each patient were used for local infiltration anesthesia (supraperiosteal injection). Injection pain was measured immediately after local infiltration anesthesia using the visual analog pain scale (VAS) questionnaire. Results from the questionnaire were used to assess degree of pain patients feel when a conventional local anesthetic technique (CNV) is used compared with a computer-controlled anesthetic delivery system (CNR). RESULTS: DAS and PSS did not correlate to injection pain. VAS scores were lower for CNR than for CNV regardless of the order in which anesthetic procedures were applied. VAS score did not differ significantly with sex. Pearson coefficient for correlation between VAS scores for the two procedures was 0.80, also indicating a strong correlation. CONCLUSION: Within the limitations of the present study, relief from injection pain is observed using CNR.

Near-unity photoluminescence quantum yield in MoS2.

Two-dimensional (2D) transition metal dichalcogenides have emerged as a promising material system for optoelectronic applications, but their primary figure of merit, the room-temperature photoluminescence quantum yield (QY), is extremely low. The prototypical 2D material molybdenum disulfide (MoS2) is reported to have a maximum QY of 0.6%, which indicates a considerable defect density. Here we report on an air-stable, solution-based chemical treatment by an organic superacid, which uniformly enhances the photoluminescence and minority carrier lifetime of MoS2 monolayers by more than two orders of magnitude. The treatment eliminates defect-mediated nonradiative recombination, thus resulting in a final QY of more than 95%, with a longest-observed lifetime of 10.8 ± 0.6 nanoseconds. Our ability to obtain optoelectronic monolayers with near-perfect properties opens the door for the development of highly efficient light-emitting diodes, lasers, and solar cells based on 2D materials.