Background: Particulate matter (PM) has been extensively observed due to its negative association with human health. Previousresearch revealed the possible negative effect of air pollutant exposure on oral health. However, the predictive model between air pollutant exposure and the prevalence of periodontitis has not been observed yet. Therefore, this study aims to propose a predictive model for the number of patients with periodontitis exposed to PM and atmospheric factors in South Korea using deep learning. Methods: This study is a retrospective cohort study utilizing secondary data from the Korean Statistical Information Service andthe Health Insurance Review and Assessment database for air pollution and the number of patients with periodontitis, respectively. Data from 2015 to 2022 were collected and consolidated every month, organized by region. Following data matching and management, the deep neural networks (DNN) model was applied, and the mean absolute percentage error (MAPE) value was calculated to ensure the accuracy of the model. Results: As we evaluated the DNN model with MAPE, the multivariate model of air pollution including exposure to PM2.5, PM10, and other atmospheric factors predict approximately 85% of the number of patients with periodontitis. The MAPE value ranged from 12.85 to 17.10 (mean±standard deviation=14.12±1.30), indicating a commendable level of accuracy. Conclusion In this study, the predictive model for the number of patients with periodontitis is developed based on air pollution,including exposure to PM2.5, PM10, and other atmospheric factors. Additionally, various relevant factors are incorporated into the developed predictive model to elucidate specific causal relationships. It is anticipated that future research will lead to the development of a more accurate model for predicting the number of patients with periodontitis.

Background: There is a need to update the cardiovascular (CV) Sequential Organ Failure Assessment (SOFA) score to reflect the current practice in sepsis. We previously proposed the modified CV SOFA score from data on blood pressure, norepinephrine equivalent dose, and lactate as gathered from emergency departments. In this study, we externally validated the modified CV SOFA score in multicenter intensive care unit (ICU) patients. Methods: A multicenter retrospective observational study was conducted on ICU patients at six hospitals in Korea. We included adult patients with sepsis who were admitted to ICUs. We compared the prognostic performance of the modified CV/total SOFA score and the original CV/total SOFA score in predicting 28-day mortality. Discrimination and calibration were evaluated using the area under the receiver operating characteristic curve (AUROC) and the calibration curve, respectively. Results: We analyzed 1,015 ICU patients with sepsis. In overall patients, the 28-day mortality rate was 31.2%. The predictive validity of the modified CV SOFA (AUROC, 0.712; 95% confidence interval [CI], 0.677–0.746; P < 0.001) was significantly higher than that of the original CV SOFA (AUROC, 0.644; 95% CI, 0.611–0.677). The predictive validity of modified total SOFA score for 28-day mortality was significantly higher than that of the original total SOFA (AUROC, 0.747 vs. 0.730; 95% CI, 0.715–0.779; P = 0.002). The calibration curve of the original CV SOFA for 28-day mortality showed poor calibration. In contrast, the calibration curve of the modified CV SOFA for 28-day mortality showed good calibration. Conclusion In patients with sepsis in the ICU, the modified SOFA score performed better than the original SOFA score in predicting 28-day mortality.

Determining blood loss [100% – RBV (%)] is challenging in the management of haemorrhagic shock. We derived an equation estimating RBV (%) via serial haematocrits (Hct1 , Hct2 ) by fixing infused crystalloid fluid volume (N) as [0.015 × body weight (g)]. Then, we validated it in vivo. Mathematically, the following estimation equation was derived: RBV (%) = 24k / [(Hct1 / Hct2 ) – 1]. For validation, nonongoing haemorrhagic shock was induced in Sprague–Dawley rats by withdrawing 20.0%–60.0% of their total blood volume (TBV) in 5.0% intervals (n = 9). Hct1 was checked after 10 min and normal saline N cc was infused over 10 min. Hct 2 was checked five minutes later. We applied a linear equation to explain RBV (%) with 1 / [(Hct1 / Hct2 ) – 1]. Seven rats losing 30.0%–60.0% of their TBV suffered shock persistently. For them, RBV (%) was updated as 5.67 / [(Hct1 / Hct2 ) – 1] + 32.8 (95% confidence interval [CI] of the slope: 3.14–8.21, p = 0.002, R2 = 0.87). On a Bland-Altman plot, the difference between the estimated and actual RBV was 0.00 ± 4.03%; the 95% CIs of the limits of agreements were included within the pre-determined criterion of validation (< 20%). For rats suffering from persistent, non-ongoing haemorrhagic shock, we derived and validated a simple equation estimating RBV (%). This enables the calculation of blood loss via information on serial haematocrits under a fixed N.Clinical validation is required before utilisation for emergency care of haemorrhagic shock.