Background/Aims: Atrial fibrillation (AF) significantly contributes to global morbidity and mortality. Paroxysmal atrial fibrillation (PAF) is particularly common among patients with cryptogenic strokes or transient ischemic attacks and has a silent nature. This study aims to develop reliable artificial intelligence (AI) algorithms to detect early signs of AF in patients with normal sinus rhythm (NSR) using a 12-lead electrocardiogram (ECG). Methods: Between 2013 and 2020, 552,372 ECG traces from 318,321 patients were collected and split into training (n = 331,422), validation (n = 110,475), and test sets (n = 110,475). Deep neural networks were then trained to predict AF onset within one month of NSR. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). An explainable AI technique was employed to identify the inference evidence underlying the predictions of deep learning models. Results: The AUROC for early diagnosis of PAF was 0.905 ± 0.007. The findings reveal that the vicinity of the T wave, including the ST segment and S-peak, significantly influences the ability of the trained neural network to diagnose PAF. Additionally, comparing the summarized ECG in NSR with those in PAF revealed that nonspecific ST-T abnormalities and inverted T waves were associated with PAF. Conclusions Deep learning can predict AF onset from NSR while detecting key features that influence decisions. This suggests that identifying undetected AF may serve as a predictive tool for PAF screening, offering valuable insights into cardiac dysfunction and stroke risk.

Background/Aims: Atrial fibrillation (AF) significantly contributes to global morbidity and mortality. Paroxysmal atrial fibrillation (PAF) is particularly common among patients with cryptogenic strokes or transient ischemic attacks and has a silent nature. This study aims to develop reliable artificial intelligence (AI) algorithms to detect early signs of AF in patients with normal sinus rhythm (NSR) using a 12-lead electrocardiogram (ECG). Methods: Between 2013 and 2020, 552,372 ECG traces from 318,321 patients were collected and split into training (n = 331,422), validation (n = 110,475), and test sets (n = 110,475). Deep neural networks were then trained to predict AF onset within one month of NSR. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). An explainable AI technique was employed to identify the inference evidence underlying the predictions of deep learning models. Results: The AUROC for early diagnosis of PAF was 0.905 ± 0.007. The findings reveal that the vicinity of the T wave, including the ST segment and S-peak, significantly influences the ability of the trained neural network to diagnose PAF. Additionally, comparing the summarized ECG in NSR with those in PAF revealed that nonspecific ST-T abnormalities and inverted T waves were associated with PAF. Conclusions Deep learning can predict AF onset from NSR while detecting key features that influence decisions. This suggests that identifying undetected AF may serve as a predictive tool for PAF screening, offering valuable insights into cardiac dysfunction and stroke risk.

Background/Aims: Atrial fibrillation (AF) significantly contributes to global morbidity and mortality. Paroxysmal atrial fibrillation (PAF) is particularly common among patients with cryptogenic strokes or transient ischemic attacks and has a silent nature. This study aims to develop reliable artificial intelligence (AI) algorithms to detect early signs of AF in patients with normal sinus rhythm (NSR) using a 12-lead electrocardiogram (ECG). Methods: Between 2013 and 2020, 552,372 ECG traces from 318,321 patients were collected and split into training (n = 331,422), validation (n = 110,475), and test sets (n = 110,475). Deep neural networks were then trained to predict AF onset within one month of NSR. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). An explainable AI technique was employed to identify the inference evidence underlying the predictions of deep learning models. Results: The AUROC for early diagnosis of PAF was 0.905 ± 0.007. The findings reveal that the vicinity of the T wave, including the ST segment and S-peak, significantly influences the ability of the trained neural network to diagnose PAF. Additionally, comparing the summarized ECG in NSR with those in PAF revealed that nonspecific ST-T abnormalities and inverted T waves were associated with PAF. Conclusions Deep learning can predict AF onset from NSR while detecting key features that influence decisions. This suggests that identifying undetected AF may serve as a predictive tool for PAF screening, offering valuable insights into cardiac dysfunction and stroke risk.

Background: Immunocompromised (IC) pediatric patients are at increased risk of severe acute respiratory syndrome coronavirus 2 infection, but the viral kinetics and seroimmunologic response in pediatric IC patients are not fully understood. Methods: From April to June 2022, a prospective cohort study was conducted. IC pediatric patients hospitalized for coronavirus disease 2019 (COVID-19) were enrolled. Serial saliva swab and serum specimens were subjected to reverse transcription polymerase chain reaction assays with mutation sequencing, viral culture, anti-spike-protein, anti-nucleocapsid antibody assays, plaque reduction neutralization test (PRNT) and multiplex cytokine assays. Results: Eleven IC children were evaluated. Their COVID-19 symptoms resolved promptly (median, 2.5 days; interquartile range, 2.0–4.3). Saliva swab specimens contained lower viral loads than nasopharyngeal swabs (P = 0.008). All cases were BA.2 infection, and 45.5% tested negative within 14 days by saliva swab from symptom onset. Eight (72.7%) showed a time-dependent increase in BA.2 PRNT titers, followed by rapid waning. Multiplex cytokine assays revealed that monocyte/macrophage activation and Th 1 responses were comparable to those of non-IC adults. Activation of interleukin (IL)-1Ra and IL-6 was brief, and IL-17A was suppressed. Activated interferon (IFN)-γ and IL-18/IL-1F4 signals were observed. Conclusion IC pediatric patients rapidly recovered from COVID-19 with low viral loads.Antibody response was limited, but cytokine analysis suggested an enhanced IFN-γ- and IL-18-mediated immune response without excessive activation of inflammatory cascades. To validate our observation, immune cell-based functional studies need to be conducted among IC and non-IC children.

Background: Immunocompromised (IC) pediatric patients are at increased risk of severe acute respiratory syndrome coronavirus 2 infection, but the viral kinetics and seroimmunologic response in pediatric IC patients are not fully understood. Methods: From April to June 2022, a prospective cohort study was conducted. IC pediatric patients hospitalized for coronavirus disease 2019 (COVID-19) were enrolled. Serial saliva swab and serum specimens were subjected to reverse transcription polymerase chain reaction assays with mutation sequencing, viral culture, anti-spike-protein, anti-nucleocapsid antibody assays, plaque reduction neutralization test (PRNT) and multiplex cytokine assays. Results: Eleven IC children were evaluated. Their COVID-19 symptoms resolved promptly (median, 2.5 days; interquartile range, 2.0–4.3). Saliva swab specimens contained lower viral loads than nasopharyngeal swabs (P = 0.008). All cases were BA.2 infection, and 45.5% tested negative within 14 days by saliva swab from symptom onset. Eight (72.7%) showed a time-dependent increase in BA.2 PRNT titers, followed by rapid waning. Multiplex cytokine assays revealed that monocyte/macrophage activation and Th 1 responses were comparable to those of non-IC adults. Activation of interleukin (IL)-1Ra and IL-6 was brief, and IL-17A was suppressed. Activated interferon (IFN)-γ and IL-18/IL-1F4 signals were observed. Conclusion IC pediatric patients rapidly recovered from COVID-19 with low viral loads.Antibody response was limited, but cytokine analysis suggested an enhanced IFN-γ- and IL-18-mediated immune response without excessive activation of inflammatory cascades. To validate our observation, immune cell-based functional studies need to be conducted among IC and non-IC children.

Background/Aims: Atrial fibrillation (AF) significantly contributes to global morbidity and mortality. Paroxysmal atrial fibrillation (PAF) is particularly common among patients with cryptogenic strokes or transient ischemic attacks and has a silent nature. This study aims to develop reliable artificial intelligence (AI) algorithms to detect early signs of AF in patients with normal sinus rhythm (NSR) using a 12-lead electrocardiogram (ECG). Methods: Between 2013 and 2020, 552,372 ECG traces from 318,321 patients were collected and split into training (n = 331,422), validation (n = 110,475), and test sets (n = 110,475). Deep neural networks were then trained to predict AF onset within one month of NSR. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). An explainable AI technique was employed to identify the inference evidence underlying the predictions of deep learning models. Results: The AUROC for early diagnosis of PAF was 0.905 ± 0.007. The findings reveal that the vicinity of the T wave, including the ST segment and S-peak, significantly influences the ability of the trained neural network to diagnose PAF. Additionally, comparing the summarized ECG in NSR with those in PAF revealed that nonspecific ST-T abnormalities and inverted T waves were associated with PAF. Conclusions Deep learning can predict AF onset from NSR while detecting key features that influence decisions. This suggests that identifying undetected AF may serve as a predictive tool for PAF screening, offering valuable insights into cardiac dysfunction and stroke risk.

Background: Immunocompromised (IC) pediatric patients are at increased risk of severe acute respiratory syndrome coronavirus 2 infection, but the viral kinetics and seroimmunologic response in pediatric IC patients are not fully understood. Methods: From April to June 2022, a prospective cohort study was conducted. IC pediatric patients hospitalized for coronavirus disease 2019 (COVID-19) were enrolled. Serial saliva swab and serum specimens were subjected to reverse transcription polymerase chain reaction assays with mutation sequencing, viral culture, anti-spike-protein, anti-nucleocapsid antibody assays, plaque reduction neutralization test (PRNT) and multiplex cytokine assays. Results: Eleven IC children were evaluated. Their COVID-19 symptoms resolved promptly (median, 2.5 days; interquartile range, 2.0–4.3). Saliva swab specimens contained lower viral loads than nasopharyngeal swabs (P = 0.008). All cases were BA.2 infection, and 45.5% tested negative within 14 days by saliva swab from symptom onset. Eight (72.7%) showed a time-dependent increase in BA.2 PRNT titers, followed by rapid waning. Multiplex cytokine assays revealed that monocyte/macrophage activation and Th 1 responses were comparable to those of non-IC adults. Activation of interleukin (IL)-1Ra and IL-6 was brief, and IL-17A was suppressed. Activated interferon (IFN)-γ and IL-18/IL-1F4 signals were observed. Conclusion IC pediatric patients rapidly recovered from COVID-19 with low viral loads.Antibody response was limited, but cytokine analysis suggested an enhanced IFN-γ- and IL-18-mediated immune response without excessive activation of inflammatory cascades. To validate our observation, immune cell-based functional studies need to be conducted among IC and non-IC children.

Background/Aims: Atrial fibrillation (AF) significantly contributes to global morbidity and mortality. Paroxysmal atrial fibrillation (PAF) is particularly common among patients with cryptogenic strokes or transient ischemic attacks and has a silent nature. This study aims to develop reliable artificial intelligence (AI) algorithms to detect early signs of AF in patients with normal sinus rhythm (NSR) using a 12-lead electrocardiogram (ECG). Methods: Between 2013 and 2020, 552,372 ECG traces from 318,321 patients were collected and split into training (n = 331,422), validation (n = 110,475), and test sets (n = 110,475). Deep neural networks were then trained to predict AF onset within one month of NSR. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). An explainable AI technique was employed to identify the inference evidence underlying the predictions of deep learning models. Results: The AUROC for early diagnosis of PAF was 0.905 ± 0.007. The findings reveal that the vicinity of the T wave, including the ST segment and S-peak, significantly influences the ability of the trained neural network to diagnose PAF. Additionally, comparing the summarized ECG in NSR with those in PAF revealed that nonspecific ST-T abnormalities and inverted T waves were associated with PAF. Conclusions Deep learning can predict AF onset from NSR while detecting key features that influence decisions. This suggests that identifying undetected AF may serve as a predictive tool for PAF screening, offering valuable insights into cardiac dysfunction and stroke risk.

Background: Immunocompromised (IC) pediatric patients are at increased risk of severe acute respiratory syndrome coronavirus 2 infection, but the viral kinetics and seroimmunologic response in pediatric IC patients are not fully understood. Methods: From April to June 2022, a prospective cohort study was conducted. IC pediatric patients hospitalized for coronavirus disease 2019 (COVID-19) were enrolled. Serial saliva swab and serum specimens were subjected to reverse transcription polymerase chain reaction assays with mutation sequencing, viral culture, anti-spike-protein, anti-nucleocapsid antibody assays, plaque reduction neutralization test (PRNT) and multiplex cytokine assays. Results: Eleven IC children were evaluated. Their COVID-19 symptoms resolved promptly (median, 2.5 days; interquartile range, 2.0–4.3). Saliva swab specimens contained lower viral loads than nasopharyngeal swabs (P = 0.008). All cases were BA.2 infection, and 45.5% tested negative within 14 days by saliva swab from symptom onset. Eight (72.7%) showed a time-dependent increase in BA.2 PRNT titers, followed by rapid waning. Multiplex cytokine assays revealed that monocyte/macrophage activation and Th 1 responses were comparable to those of non-IC adults. Activation of interleukin (IL)-1Ra and IL-6 was brief, and IL-17A was suppressed. Activated interferon (IFN)-γ and IL-18/IL-1F4 signals were observed. Conclusion IC pediatric patients rapidly recovered from COVID-19 with low viral loads.Antibody response was limited, but cytokine analysis suggested an enhanced IFN-γ- and IL-18-mediated immune response without excessive activation of inflammatory cascades. To validate our observation, immune cell-based functional studies need to be conducted among IC and non-IC children.

While the Korean Triage and Acuity Scale (KTAS) was introduced in 2016 as a tool to identify patients at risk of catastrophic events, including death in the ED, the triage system for the pre-hospital stage still lacks evidence. The pre-hospital stage is characterized by time-sensitive and complex scenarios, where rapid and accurate decision-making is paramount to optimize patient outcomes. Despite the vital role of pre-hospital care providers, the invalidated and subjective current triage system consisting of 4-stages is still used at the pre-hospital stage, and hence, it needs to be modified to be more objective, standardized, and reliable. To improve the Korean emergency medical system, the pre-hospital KTAS (Pre-KTAS) was developed in 2020, and then two pilot projects were conducted in 2022 and 2023. This paper not only reveals the results of the first and second pilot projects for Pre-KTAS but also highlights the potential benefits of using this newly developed triage tool in the pre-hospital setting. Furthermore, this paper suggests ways to improve the emergency medical system (EMS) in Korea by improving patient safety, resource allocation, and overall emergency response efficiency.