Purpose: Early identification of patients at risk for acute respiratory failure (ARF) could help clinicians devise preventive strategies. Analyzing biosignals with artificial intelligence (AI) can uncover hidden information and variability within time series. We aimed to develop and validate AI models to predict ARF within 72 h after emergency department admission, primarily using highresolution biosignals collected within 4 h of arrival. Materials and Methods: Our AI model, built on convolutional recurrent neural networks, combines biosignal feature extraction and sequence modeling. The model was developed and internally validated with data from 5284 admissions [1085 (20.5%) positive for ARF], and externally validated using data from 144 admissions [7 (4.9%) positive for ARF] from another institution. We defined ARF as the application of advanced respiratory support devices. Results: Our AI model performed well in predicting ARF, achieving area under the receiver operating characteristic curve (AUROC) of 0.840 and 0.743 in internal and external validations, respectively. It outperformed the Modified Early Warning Score (MEWS) and XGBoost models built only with clinical variables. High predictive ability for mortality was observed, with AUROC up to 0.809. A 10% increase in AI prediction scores was associated with 1.44-fold and 1.42-fold increases in ARF risk and mortality risk, respectively, even after adjusting for MEWS and demographic variables. Conclusion Our AI model demonstrates high predictive accuracy and significant associations with clinical outcomes. Our AI model has the potential to promptly aid in triage decisions. Our study shows that using AI to analyze biosignals advances disease detection and prediction.

Purpose: Early identification of patients at risk for acute respiratory failure (ARF) could help clinicians devise preventive strategies. Analyzing biosignals with artificial intelligence (AI) can uncover hidden information and variability within time series. We aimed to develop and validate AI models to predict ARF within 72 h after emergency department admission, primarily using highresolution biosignals collected within 4 h of arrival. Materials and Methods: Our AI model, built on convolutional recurrent neural networks, combines biosignal feature extraction and sequence modeling. The model was developed and internally validated with data from 5284 admissions [1085 (20.5%) positive for ARF], and externally validated using data from 144 admissions [7 (4.9%) positive for ARF] from another institution. We defined ARF as the application of advanced respiratory support devices. Results: Our AI model performed well in predicting ARF, achieving area under the receiver operating characteristic curve (AUROC) of 0.840 and 0.743 in internal and external validations, respectively. It outperformed the Modified Early Warning Score (MEWS) and XGBoost models built only with clinical variables. High predictive ability for mortality was observed, with AUROC up to 0.809. A 10% increase in AI prediction scores was associated with 1.44-fold and 1.42-fold increases in ARF risk and mortality risk, respectively, even after adjusting for MEWS and demographic variables. Conclusion Our AI model demonstrates high predictive accuracy and significant associations with clinical outcomes. Our AI model has the potential to promptly aid in triage decisions. Our study shows that using AI to analyze biosignals advances disease detection and prediction.

Purpose: Early identification of patients at risk for acute respiratory failure (ARF) could help clinicians devise preventive strategies. Analyzing biosignals with artificial intelligence (AI) can uncover hidden information and variability within time series. We aimed to develop and validate AI models to predict ARF within 72 h after emergency department admission, primarily using highresolution biosignals collected within 4 h of arrival. Materials and Methods: Our AI model, built on convolutional recurrent neural networks, combines biosignal feature extraction and sequence modeling. The model was developed and internally validated with data from 5284 admissions [1085 (20.5%) positive for ARF], and externally validated using data from 144 admissions [7 (4.9%) positive for ARF] from another institution. We defined ARF as the application of advanced respiratory support devices. Results: Our AI model performed well in predicting ARF, achieving area under the receiver operating characteristic curve (AUROC) of 0.840 and 0.743 in internal and external validations, respectively. It outperformed the Modified Early Warning Score (MEWS) and XGBoost models built only with clinical variables. High predictive ability for mortality was observed, with AUROC up to 0.809. A 10% increase in AI prediction scores was associated with 1.44-fold and 1.42-fold increases in ARF risk and mortality risk, respectively, even after adjusting for MEWS and demographic variables. Conclusion Our AI model demonstrates high predictive accuracy and significant associations with clinical outcomes. Our AI model has the potential to promptly aid in triage decisions. Our study shows that using AI to analyze biosignals advances disease detection and prediction.

Purpose: Early identification of patients at risk for acute respiratory failure (ARF) could help clinicians devise preventive strategies. Analyzing biosignals with artificial intelligence (AI) can uncover hidden information and variability within time series. We aimed to develop and validate AI models to predict ARF within 72 h after emergency department admission, primarily using highresolution biosignals collected within 4 h of arrival. Materials and Methods: Our AI model, built on convolutional recurrent neural networks, combines biosignal feature extraction and sequence modeling. The model was developed and internally validated with data from 5284 admissions [1085 (20.5%) positive for ARF], and externally validated using data from 144 admissions [7 (4.9%) positive for ARF] from another institution. We defined ARF as the application of advanced respiratory support devices. Results: Our AI model performed well in predicting ARF, achieving area under the receiver operating characteristic curve (AUROC) of 0.840 and 0.743 in internal and external validations, respectively. It outperformed the Modified Early Warning Score (MEWS) and XGBoost models built only with clinical variables. High predictive ability for mortality was observed, with AUROC up to 0.809. A 10% increase in AI prediction scores was associated with 1.44-fold and 1.42-fold increases in ARF risk and mortality risk, respectively, even after adjusting for MEWS and demographic variables. Conclusion Our AI model demonstrates high predictive accuracy and significant associations with clinical outcomes. Our AI model has the potential to promptly aid in triage decisions. Our study shows that using AI to analyze biosignals advances disease detection and prediction.

Purpose: Early identification of patients at risk for acute respiratory failure (ARF) could help clinicians devise preventive strategies. Analyzing biosignals with artificial intelligence (AI) can uncover hidden information and variability within time series. We aimed to develop and validate AI models to predict ARF within 72 h after emergency department admission, primarily using highresolution biosignals collected within 4 h of arrival. Materials and Methods: Our AI model, built on convolutional recurrent neural networks, combines biosignal feature extraction and sequence modeling. The model was developed and internally validated with data from 5284 admissions [1085 (20.5%) positive for ARF], and externally validated using data from 144 admissions [7 (4.9%) positive for ARF] from another institution. We defined ARF as the application of advanced respiratory support devices. Results: Our AI model performed well in predicting ARF, achieving area under the receiver operating characteristic curve (AUROC) of 0.840 and 0.743 in internal and external validations, respectively. It outperformed the Modified Early Warning Score (MEWS) and XGBoost models built only with clinical variables. High predictive ability for mortality was observed, with AUROC up to 0.809. A 10% increase in AI prediction scores was associated with 1.44-fold and 1.42-fold increases in ARF risk and mortality risk, respectively, even after adjusting for MEWS and demographic variables. Conclusion Our AI model demonstrates high predictive accuracy and significant associations with clinical outcomes. Our AI model has the potential to promptly aid in triage decisions. Our study shows that using AI to analyze biosignals advances disease detection and prediction.

Background: Hot springs have been traditionally used as an alternative treatment for a wide range of diseases, including rheumatoid arthritis, bronchial asthma, diabetes, hypertension, psoriasis and atopic dermatitis. However, the clinical effects and therapeutic mechanisms associated with hot springs remain poorly defined. Objective: The purpose of this study was to demonstrate the different effects of hot springs on cellular viability and secretion of inflammatory cytokines on keratinocyte in two geographically representative types of hot springs: NaHCO3 -type and NaCl-type, which are the most common types in South Korea. Methods: We performed WST-1, BrdU measurements, human inflammatory cytokine arrays and enzyme-linked immunosorbent assay in HaCaT cells stimulated with toll-like receptor 3 by polyinosinic-polycytidylic acid. Results: The interaction effects of cell viability and cell proliferation were not significantly different regardless of polyinosinic-polycytidylic acid stimulation and cultured hot springs type. Cytokine array and enzyme-linked immunosorbent assay analysis showed increased expression of inflammatory cytokines such as interleukin-6 and granulocyte-macrophage colony-stimulating factor by polyinosinic-polycytidylic acid stimulation, with expression levels differing according to hot springs hydrochemical composition. Cytokine reduction was not significant. Conclusion The effects and mechanisms of hot springs treatment in keratinocytes were partially elucidated.

Background: Hot springs have been traditionally used as an alternative treatment for a wide range of diseases, including rheumatoid arthritis, bronchial asthma, diabetes, hypertension, psoriasis and atopic dermatitis. However, the clinical effects and therapeutic mechanisms associated with hot springs remain poorly defined. Objective: The purpose of this study was to demonstrate the different effects of hot springs on cellular viability and secretion of inflammatory cytokines on keratinocyte in two geographically representative types of hot springs: NaHCO3 -type and NaCl-type, which are the most common types in South Korea. Methods: We performed WST-1, BrdU measurements, human inflammatory cytokine arrays and enzyme-linked immunosorbent assay in HaCaT cells stimulated with toll-like receptor 3 by polyinosinic-polycytidylic acid. Results: The interaction effects of cell viability and cell proliferation were not significantly different regardless of polyinosinic-polycytidylic acid stimulation and cultured hot springs type. Cytokine array and enzyme-linked immunosorbent assay analysis showed increased expression of inflammatory cytokines such as interleukin-6 and granulocyte-macrophage colony-stimulating factor by polyinosinic-polycytidylic acid stimulation, with expression levels differing according to hot springs hydrochemical composition. Cytokine reduction was not significant. Conclusion The effects and mechanisms of hot springs treatment in keratinocytes were partially elucidated.

BACKGROUND: Empirical evidences for efficacy of hot spring (HS) water in inflammatory skin disorders have not been substantiated with sufficient, immunological “hard evidence”. Mageumsan HS water, characterized by its weakly-alkaline properties and low total dissolved solids content, has been known to alleviate various immune-inflammatory skin diseases, including atopic dermatitis (AD). OBJECTIVE: The trial attempted to quantitatively analyze in vitro expression levels of chemical mediators in cutaneous inflammation from HaCaT cell line treated with Mageumsan HS, and suggest the likely mode of action through which it exerts the apparent anti-inflammatory effects in AD. METHODS: Using membrane-based human antibody array kit, customized to include 30 different, keratinocyte-derived mediator proteins, their expression levels (including interleukin [IL]-1, IL-6, IL-8, thymic stromal lymphopoietin, thymus and activation-regulated chemokine, and granulocyte macrophage colony-stimulating factor) were assessed in vitro. Selected key proteins were further quantified with enzyme-linked immunosorbent assay. RESULTS: There was a clear pattern of overall suppression of the mediators, especially those noted for their pro-inflammatory role in AD (monocyte chemoattractant protein [MCP]-1, regulated on activation, normal T cell expressed and secreted, cutaneous T-cell-attracting chemokine, Eotaxin, and macrophage inflammatory protein-1α, etc.). Also, reduced expression of involucrin and cytokeratin 1 was also reduced in the HS-treated group. CONCLUSION: The present study has shown that Mageumsan HS water may exert its effects on inflammatory skin disorders through regulation of proinflammatory cytokines. These evidences are to be supported with further future investigations to elucidate immunological mechanism behind these beneficial effects of HS water in the chronically inflamed skin of AD.
Cell Line ; Chemokine CCL17 ; Chemokine CCL27 ; Cytokines ; Dermatitis, Atopic ; Enzyme-Linked Immunosorbent Assay ; Granulocytes ; Hot Springs* ; Humans ; In Vitro Techniques* ; Inflammation ; Interleukin-6 ; Interleukin-8 ; Interleukins ; Keratins ; Macrophages ; Protein Array Analysis* ; Skin ; Skin Diseases ; Water

BACKGROUND: Human mannose-binding lectin (MBL) is a serum lectin taking part in the innate immunity by opsonizing various microorganisms for phagocytosis. The MBL serum concentration is affected by several single-nucleotide polymorphisms (SNPs) in the promoter region of the MBL2 gene. OBJECTIVE: The purpose of this study was to examine the relationship between MBL2 polymorphisms and atopic dermatitis (AD) susceptibility. METHODS: To examine whether the MBL2 SNPs are related to AD susceptibility, we examined 237 patients with AD and 94 controls by polymerase chain reaction (PCR)-restriction fragment length polymorphism and PCR-sequence specific primer analyses of four polymorphic loci: two (H/L and X/Y) within the promoter region and the other two (P/Q and A/B) within exon 1. MBL concentrations in the blood were estimated by ELISA. RESULTS: The prevalence of haplotype HYPB, leading to MBL deficiency, was significantly decreased in the AD patients compared to the controls (p=0.002), while the prevalence of haplotype HYPA was increased with a clear trend toward significance (p=0.056). The frequency of MBL2 LYPB/LXPA (odds ratio, 0.08; 95% confidence interval, 0.009~0.655; p=0.021) were significantly decreased in the AD patients. The blood log [total immunoglobulin E, IgE] levels of MBL2 HYPA/HYPA, HYPA/LYPA, HYPA/LYPB, HYPA/LYQA, and LYQA/LXPA haplotype pairs were significantly increased in the AD patients. CONCLUSION: The frequency of MBL2 HYPB haplotype was significantly decreased in the AD patients compared to the controls. The frequency of LYPB/LXPA had a possibly protective effect on AD. Moreover, the MBL2 HYPA haplotype pairs, which were related to higher blood total IgE levels, were possibly associated with extrinsic AD.
Dermatitis, Atopic* ; Enzyme-Linked Immunosorbent Assay ; Exons ; Haplotypes ; Humans ; Immunity, Innate ; Immunoglobulin E ; Immunoglobulins ; Mannose-Binding Lectin ; Phagocytosis ; Polymerase Chain Reaction ; Polymorphism, Single Nucleotide ; Prevalence ; Promoter Regions, Genetic

No abstract available.
Acne Vulgaris* ; Humans ; Pilot Projects* ; Skin*