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: 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/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.

Purpose: Biliary atresia (BA) is a rare but progressive cholangiopathy and the leading cause of pediatric liver transplantation worldwide. The Korean Association of Pediatric Surgeons (KAPS) has conducted three national surveys (2001, 2011, and 2023) to assess long-term trends in the diagnosis, treatment, and outcomes of BA. This study provides a comparative analysis of the 2nd and 3rd national surveys, with reference to selected findings from the 1st survey. Methods: This study included 453 patients from the 3rd national survey (2011–2021) and 435 patients from the 2nd survey (2001–2010), all of whom underwent Kasai portoenterostomy. Data were collected via electronic case report forms from pediatric surgical centers nationwide. Comparisons were made regarding demographics, clinical features, diagnostic patterns, operative details, follow-up outcomes, and survival. Kaplan–Meier analysis was used to evaluate long-term survival. Results: The mean number of BA patients per year remained stable between surveys (43.5 in the 2nd, 41.18 in the 3rd), though centralization of care increased, with 61.5% of cases managed by two major institutions in the 3rd survey. The median age at surgery decreased, and the use of preoperative imaging (especially magnetic resonance cholangiopancreatography) increased. The 10-year native liver survival rate declined from 59.8% to 53.7%, while overall 10-year survival improved slightly (92.9% to 93.2%). Postoperative complications, such as cholangitis and liver failure, persisted but were better categorized. The 3rd survey also reported improved mortality (4.9%) and reduced follow-up loss (11.5%) compared to the 2nd survey. Conclusion While overall survival after Kasai operation has remained high and even improved, native liver survival has slightly declined. The findings reflect earlier diagnosis, more consistent diagnostic imaging, and increasing centralization of care. These trends underscore the importance of long-term nationwide data collection in guiding future strategies for BA management in Korea.