An Artificial Intelligence Model for Predicting Trauma Mortality Among Emergency Department Patients in South Korea: Retrospective Cohort Study.

BACKGROUND: Within the trauma system, the emergency department (ED) is the hospital's first contact and is vital for allocating medical resources. However, there is generally limited information about patients that die in the ED. OBJECTIVE: The aim of this study was to develop an artificial intelligence (AI) model to predict trauma mortality and analyze pertinent mortality factors for all patients visiting the ED. METHODS: We used the Korean National Emergency Department Information System (NEDIS) data set (N=6,536,306), incorporating over 400 hospitals between 2016 and 2019. We included the International Classification of Disease 10th Revision (ICD-10) codes and chose the following input features to predict ED patient mortality: age, sex, intentionality, injury, emergent symptom, Alert/Verbal/Painful/Unresponsive (AVPU) scale, Korean Triage and Acuity Scale (KTAS), and vital signs. We compared three different feature set performances for AI input: all features (n=921), ICD-10 features (n=878), and features excluding ICD-10 codes (n=43). We devised various machine learning models with an ensemble approach via 5-fold cross-validation and compared the performance of each model with that of traditional prediction models. Lastly, we investigated explainable AI feature effects and deployed our final AI model on a public website, providing access to our mortality prediction results among patients visiting the ED. RESULTS: Our proposed AI model with the all-feature set achieved the highest area under the receiver operating characteristic curve (AUROC) of 0.9974 (adaptive boosting [AdaBoost], AdaBoost + light gradient boosting machine [LightGBM]: Ensemble), outperforming other state-of-the-art machine learning and traditional prediction models, including extreme gradient boosting (AUROC=0.9972), LightGBM (AUROC=0.9973), ICD-based injury severity scores (AUC=0.9328 for the inclusive model and AUROC=0.9567 for the exclusive model), and KTAS (AUROC=0.9405). In addition, our proposed AI model outperformed a cutting-edge AI model designed for in-hospital mortality prediction (AUROC=0.7675) for all ED visitors. From the AI model, we also discovered that age and unresponsiveness (coma) were the top two mortality predictors among patients visiting the ED, followed by oxygen saturation, multiple rib fractures (ICD-10 code S224), painful response (stupor, semicoma), and lumbar vertebra fracture (ICD-10 code S320). CONCLUSIONS: Our proposed AI model exhibits remarkable accuracy in predicting ED mortality. Including the necessity for external validation, a large nationwide data set would provide a more accurate model and minimize overfitting. We anticipate that our AI-based risk calculator tool will substantially aid health care providers, particularly regarding triage and early diagnosis for trauma patients.

Model for Predicting In-Hospital Mortality of Physical Trauma Patients Using Artificial Intelligence Techniques: Nationwide Population-Based Study in Korea.

BACKGROUND: Physical trauma-related mortality places a heavy burden on society. Estimating the mortality risk in physical trauma patients is crucial to enhance treatment efficiency and reduce this burden. The most popular and accurate model is the Injury Severity Score (ISS), which is based on the Abbreviated Injury Scale (AIS), an anatomical injury severity scoring system. However, the AIS requires specialists to code the injury scale by reviewing a patient's medical record; therefore, applying the model to every hospital is impossible. OBJECTIVE: We aimed to develop an artificial intelligence (AI) model to predict in-hospital mortality in physical trauma patients using the International Classification of Disease 10th Revision (ICD-10), triage scale, procedure codes, and other clinical features. METHODS: We used the Korean National Emergency Department Information System (NEDIS) data set (N=778,111) compiled from over 400 hospitals between 2016 and 2019. To predict in-hospital mortality, we used the following as input features: ICD-10, patient age, gender, intentionality, injury mechanism, and emergent symptom, Alert/Verbal/Painful/Unresponsive (AVPU) scale, Korean Triage and Acuity Scale (KTAS), and procedure codes. We proposed the ensemble of deep neural networks (EDNN) via 5-fold cross-validation and compared them with other state-of-the-art machine learning models, including traditional prediction models. We further investigated the effect of the features. RESULTS: Our proposed EDNN with all features provided the highest area under the receiver operating characteristic (AUROC) curve of 0.9507, outperforming other state-of-the-art models, including the following traditional prediction models: Adaptive Boosting (AdaBoost; AUROC of 0.9433), Extreme Gradient Boosting (XGBoost; AUROC of 0.9331), ICD-based ISS (AUROC of 0.8699 for an inclusive model and AUROC of 0.8224 for an exclusive model), and KTAS (AUROC of 0.1841). In addition, using all features yielded a higher AUROC than any other partial features, namely, EDNN with the features of ICD-10 only (AUROC of 0.8964) and EDNN with the features excluding ICD-10 (AUROC of 0.9383). CONCLUSIONS: Our proposed EDNN with all features outperforms other state-of-the-art models, including the traditional diagnostic code-based prediction model and triage scale.

Tunable Kondo Resonance at a Pristine Two-Dimensional Dirac Semimetal on a Kondo Insulator.

The proximity of two different materials leads to an intricate coupling of quasiparticles so that an unprecedented electronic state is often realized at the interface. Here, we demonstrate a resonance-type many-body ground state in graphene, a nonmagnetic two-dimensional Dirac semimetal, when grown on SmB6, a Kondo insulator, via thermal decomposition of fullerene molecules. This ground state is typically observed in three-dimensional magnetic materials with correlated electrons. Above the characteristic Kondo temperature of the substrate, the electron band structure of pristine graphene remains almost intact. As temperature decreases, however, the Dirac Fermions of graphene become hybridized with the Sm 4f states. Remarkable enhancement of the hybridization and Kondo resonance is observed with further cooling and increasing charge-carrier density of graphene, evidencing the Kondo screening of the Sm 4f local magnetic moment by the conduction electrons of graphene at the interface. These findings manifest the realization of the Kondo effect in graphene by the proximity of SmB6 that is tuned by the temperature and charge-carrier density of graphene.

Association between Near Work, Physical Activities and Myopia in Korean Adults During COVID-19 Outbreak.

PURPOSE: This study investigated the relationship between near work hours and myopia in Korean adults. METHODS: We used data from the 2021 Korean National Health and Nutrition Examination Survey. Associations between near work time, physical activity, and myopia were assessed using chi-square tests and multiple logistic regression analyses. RESULTS: The overall prevalence of myopia was 60.2% in adults aged 19-59 years. The prevalence of myopia was 46.2% for individuals who used smart devices for less than one hour per day, while it was 68.0% for those who used smart devices for more than four hours. In the multiple logistic regression analysis, the odds ratio (OR) for myopia was significantly higher among individuals using smart devices for 3 hours (OR = 1.55, 95% CI = 1.08-2.23) or more than 4 hours (OR = 1.75, 95% CI = 1.27-2.42), compared to users with less than 1 hour of usage. Regarding sitting time, the OR for myopia was significantly higher in individuals who sat for more than 12 hours (OR = 1.66, 95% CI = 1.05-2.61) compared to those who sat less than 4 hours. CONCLUSION: This study found that near work and sitting times were positively associated with myopia. Given the high prevalence of myopia and its implications for serious eye diseases, it is essential to implement measures to manage myopia. Considering the increased near work hours resulting from the COVID-19 pandemic, it is necessary to adopt supplementary measures, such as ensuring sufficient rest time for the eyes and adjusting the brightness of lights, to improve eye health.

Simultaneous determination of sildenafil and its active metabolite UK-103,320 in human plasma using liquid chromatography-tandem mass spectrometry.

A liquid chromatography-tandem mass spectrometric method for the simultaneous determination of sildenafil and its active N-demethylated metabolite, UK-103,320 in human plasma was developed. Sildenafil, UK-103,320 and the internal standard (DA-8159) were extracted from human plasma with dichloromethane at basic pH. A reverse-phase LC separation was performed on Luna phenylhexyl column with the mixture of acetonitrile-ammonium formate (10 mM, pH 6.0) (60:40, v/v) as mobile phase. The detection of analytes was performed using an electrospray ionization tandem mass spectrometry in the multiple reaction-monitoring mode. The lower limits of quantification for sildenafil and UK-103,320 were 2.0 ng/ml. The method showed a satisfactory sensitivity, precision, accuracy, recovery and selectivity.

In vitro metabolism of a new neuroprotective agent, KR-31543 in the human liver microsomes: identification of human cytochrome P450.

KR-31543, (2S,3R,4S)-6-amino-4-[N-(4-chlorophenyl)-N-(2-methyl-2H-tetrazol-5-ylmethyl)amino]-3,4-dihydro-2-dimethoxymethyl-3-hydroxy-2-methyl-2H-1-benzopyran, is a new neuroprotective agent for preventing ischemia-reperfusion damage. This study was performed to identify the metabolic pathway of KR-31543 in human liver microsomes and to characterize cytochrome P450 (CYP) enzymes that are involved in the metabolism of KR-31543. Human liver microsomal incubation of KR-31543 in the presence of NADPH resulted in the formation of two metabolites, M1 and M2. M1 was identified as N-(4-chlorophenyl)-N-(2-methyl-2H-tetrazol-5-ylmethyl)amine on the basis of LC/MS/MS analysis with a synthesized authentic standard, and M2 was suggested to be hydroxy-KR-31543. Correlation analysis between the known CYP enzyme activities and the rates of the formation of M1 and M2 in the 12 human liver microsomes have showed significant correlations with testosterone 6beta-hydroxylase activity (a marker of CYP3A4). Ketoconazole, a selective inhibitor of CYP3A4, and anti-CYP3A4 monoclonal antibodies potently inhibited both N-hydrolysis and hydroxylation of KR-31543 in human liver microsomes. These results provide evidence that CYP3A4 is the major isozyme responsible for the metabolism of KR-31543 to M1 and M2.

Metabolism of a new neuroprotective agent for ischemia-reperfusion damage, KR-31543 in the rat using liquid chromatography/electrospray mass spectrometry.

KR-31543, (2S,3R,4S)-6-amino-4-[N-(4-chlorophenyl)-N-(2-methyl-2H-tetrazol-5-ylmethyl)amino]-3,4-dihydro-2-dimethoxymethyl-3-hydroxy-2-methyl-2H-1-benzopyran is a new neuroprotective agent for ischemia-reperfusion damage. The in vitro and in vivo metabolism of KR-31543 in rats has been studied by LC-electrospray mass spectrometry. Rat liver microsomal incubation of KR-31543 in the presence of NADPH resulted in the formation of a metabolite M1. M1 was identified as N-(4-chlorophenyl)-N-(2-methyl-2H-tetrazol-5-ylmethyl)amine on the basis of LC-MS/MS analysis with the synthesized authentic standard. Rat CYP3A1 and 3A2 are the major CYP isozymes involved in the formation of M1.