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.

OBJECTIVES: Biosignal data include important physiological information. For that reason, many devices and systems have been developed, but there has not been enough consideration of how to collect and integrate raw data from multiple systems. To overcome this limitation, we have developed a system for collecting and integrating biosignal data from two patient monitoring systems. METHODS: We developed an interface to extract biosignal data from Nihon Kohden and Philips monitoring systems. The Nihon Kohden system has a central server for the temporary storage of raw waveform data, which can be requested using the HL7 protocol. However, the Philips system used in our hospital cannot save raw waveform data. Therefore, our system was connected to monitoring devices using the RS232 protocol. After collection, the data were transformed and stored in a unified format. RESULTS: From September 2016 to August 2017, we collected approximately 117 patient-years of waveform data from 1,268 patients in 79 beds of five intensive care units. Because the two systems use the same data storage format, the application software could be run without compatibility issues. CONCLUSIONS: Our system collects biosignal data from different systems in a unified format. The data collected by the system can be used to develop algorithms or applications without the need to consider the source of the data.
Electrocardiography ; Humans ; Information Storage and Retrieval ; Intensive Care Units ; Monitoring, Physiologic* ; Photoplethysmography

BACKGROUND: Proper education regarding inhaler usage and optimal management of chronic obstructive pulmonary disease (COPD) is essential for effectively treating patients with COPD. This study was conducted to evaluate the effects of a comprehensive education program including inhaler training and COPD management. METHODS: We enlisted 127 patients with COPD on an outpatient basis at 43 private clinics in Korea. The patients were educated on inhaler usage and disease management for three visits across 2 weeks. Physicians and patients were administered a COPD assessment test (CAT) and questionnaires about the correct usage of inhalers and management of COPD before commencement of this program and after their third visit. RESULTS: The outcomes of 127 COPD patients were analyzed. CAT scores (19.6±12.5 vs. 15.1±12.3) improved significantly after this program (p<0.05). Patients with improved CAT scores of 4 points or more had a better understanding of COPD management and the correct technique for using inhalers than those who did not have improved CAT scores (p<0.05). CONCLUSION: A comprehensive education program including inhaler training and COPD management at a primary care setting improved CAT scores and led to patients' better understanding of COPD management.
Animals ; Cats ; Disease Management* ; Dry Powder Inhalers ; Education* ; Humans ; Korea ; Metered Dose Inhalers ; Nebulizers and Vaporizers* ; Outpatients ; Primary Health Care ; Pulmonary Disease, Chronic Obstructive*

Acknowledgements section was missing. The publisher apologises for these errors.

Lung cancer is a leading cause of cancer-related death in the world. Smoking is definitely the most important risk factor for lung cancer. Radon (222Rn) is a natural gas produced from radium (226Ra) in the decay series of uranium (238U). Radon exposure is the second most common cause of lung cancer and the first risk factor for lung cancer in never-smokers. Case–control studies have provided epidemiological evidence of the causative relationship between indoor radon exposure and lung cancer. Twenty-four case–control study papers were found by our search strategy from the PubMed database. Among them, seven studies showed that indoor radon has a statistically significant association with lung cancer. The studies performed in radon-prone areas showed a more positive association between radon and lung cancer. Reviewed papers had inconsistent results on the dose–response relationship between indoor radon and lung cancer risk. Further refined case–control studies will be required to evaluate the relationship between radon and lung cancer. Sufficient study sample size, proper interview methods, valid and precise indoor radon measurement, wide range of indoor radon, and appropriate control of confounders such as smoking status should be considered in further case–control studies.
Lung Neoplasms* ; Lung* ; Natural Gas ; Radium ; Radon* ; Risk Factors* ; Sample Size ; Smoke ; Smoking ; Uranium

BACKGROUND: Apoptosis plays a role in the development of pleural effusion. Caspase-cleaved cytokeratin 18, a marker for epithelial cell apoptosis, was evaluated in pleural effusion. METHODS: A total of 79 patients with pleural effusion were enrolled. The underlying causes were lung cancer (n=24), parapneumonic effusion (n=15), tuberculous effusion (n=28), and transudates (n=12). The levels of M30, an epitope of caspase-cleaved cytokeratin 18, were measured in blood and pleural fluids using enzyme-linked immunosorbent assay along with routine cellular and biochemical parameters. The expression of M30 was evaluated in the pleural tissues using immunohistochemistry for M30. RESULTS: The M30 levels in pleural fluid were significantly higher in patients with tuberculosis (2,632.1+/-1,467.3 U/mL) than in patients with lung cancer (956.5+/-618.5 U/mL), parapneumonic effusion (689.9+/-413.6 U/mL), and transudates (273.6+/-144.5 U/mL; all p<0.01). The serum levels were not significantly different among the disease groups. Based on receiver operating characteristics analysis, the area under the curve of M30 for differentiating tuberculous pleural effusion from all other effusions was 0.93. In the immunohistochemical analysis of M30, all pathologic types of cancer cells showed moderate to high expression, and the epithelioid cells in granulomas showed high expression in tuberculous pleural tissues. CONCLUSION: Caspase-cleaved cytokeratin 18 was most prominently observed in tuberculous pleural effusion and showed utility as a clinical marker. The main source of M30 was found to be the epithelioid cells of granulomas in tuberculous pleural tissues.
Apoptosis ; Biomarkers ; Cytoskeleton ; Enzyme-Linked Immunosorbent Assay ; Epithelial Cells ; Epithelioid Cells ; Exudates and Transudates ; Granuloma ; Humans ; Immunohistochemistry ; Keratin-18* ; Keratins* ; Lung Neoplasms ; Pleural Effusion* ; ROC Curve ; Tuberculosis ; Tuberculosis, Pleural