Prediction of hospital-acquired influenza using machine learning algorithms: a comparative study.

BACKGROUND: Hospital-acquired influenza (HAI) is under-recognized despite its high morbidity and poor health outcomes. The early detection of HAI is crucial for curbing its transmission in hospital settings. AIM: This study aimed to investigate factors related to HAI, develop predictive models, and subsequently compare them to identify the best performing machine learning algorithm for predicting the occurrence of HAI. METHODS: This retrospective observational study was conducted in 2022 and included 111 HAI and 73,748 non-HAI patients from the 2011-2012 and 2019-2020 influenza seasons. General characteristics, comorbidities, vital signs, laboratory and chest X-ray results, and room information within the electronic medical record were analysed. Logistic Regression (LR), Random Forest (RF), Extreme Gradient Boosting (XGB), and Artificial Neural Network (ANN) techniques were used to construct the predictive models. Employing randomized allocation, 80% of the dataset constituted the training set, and the remaining 20% comprised the test set. The performance of the developed models was assessed using metrics such as the area under the receiver operating characteristic curve (AUC), the count of false negatives (FN), and the determination of feature importance. RESULTS: Patients with HAI demonstrated notable differences in general characteristics, comorbidities, vital signs, laboratory findings, chest X-ray result, and room status compared to non-HAI patients. Among the developed models, the RF model demonstrated the best performance taking into account both the AUC (83.3%) and the occurrence of FN (four). The most influential factors for prediction were staying in double rooms, followed by vital signs and laboratory results. CONCLUSION: This study revealed the characteristics of patients with HAI and emphasized the role of ventilation in reducing influenza incidence. These findings can aid hospitals in devising infection prevention strategies, and the application of machine learning-based predictive models especially RF can enable early intervention to mitigate the spread of influenza in healthcare settings.

Maternal Psychosocial Factors that Affect Breastfeeding Adaptation and Immune Substances in Human Milk.

PURPOSE: This study was to identify relationships of maternal psychosocial factors including mother's mood state, childcare stress, social support and sleep satisfaction with breastfeeding adaptation and immune substances in breast milk, especially secretory immunoglobulin A (sIgA) and transforming growth factor-beta 2 (TGF-beta2). METHODS: Data were collected from 84 mothers who delivered full-term infants by natural childbirth. Structured questionnaires and breast milk were collected at 2~4 days and 6 weeks postpartum. Data were analyzed using descriptive statistics, Pearson's correlation, multiple linear regression, and generalized estimating equation (GEE). RESULTS: Scores for the breastfeeding adaptation scale were significantly related with child care stress, mood state and social support. Mother's anger was positively correlated with the level of sIgA in colostrum (p<.01). Immune substances of breastmilk was significantly influenced by time for milk collection (p<.001) and the type of breastfeeding (sIgA, p<.001, TGF-beta2, p=.003). Regression analysis showed that breastfeeding adaptation could be explained 59.1% by the type of breastfeeding, childcare stress, the Profile of Mood States, emotional support and sleep quality (F=16.67, p<.001). CONCLUSION: The findings from this study provide important concepts of breastfeeding adaptation program and explanation of psychosocial factors by immune substances in breast milk. Future research, specially, bio-maker research on breast milk should focus on the ways to improve breastfeeding adaptation.

Cu/Zn incorporation during purification of soluble human EC-SOD from E. coli stabilizes proper disulfide bond formation.

Extracellular superoxide dismutase (EC-SOD) is the only enzyme that removes superoxide radical in the extracellular space. The reduction of EC-SOD is linked to many diseases, suggesting that the protein may have therapeutic value. EC-SOD is reported to be insoluble and to make inclusion bodies when overexpressed in the cytoplasm of Escherichia coli. The refolding process has the advantage of high yield, but has the disadvantage of frequent aggregation or misfolding during purification. For the first time, this study shows that fusion with maltose-binding protein (MBP), N-utilization substance protein A, and protein disulfide isomerase enabled the soluble overexpression of EC-SOD in the cytoplasm of E. coli. MBP-tagged human EC-SOD (hEC-SOD) was purified by MBP affinity and anion exchange chromatography, and its identity was confirmed by MALDI-TOF MS analysis. The purified protein showed good enzyme activity in vitro; however, there was a difference in metal binding. When copper and zinc were incorporated into hEC-SOD before MBP tag cleavage, the enzymatic activity was higher than when the metal ions were bound to the purified protein after MBP tag cleavage. Therefore, the enzymatic activity of hEC-SOD is associated with metal incorporation and protein folding via disulfide bond.

Serum carcinoembryonic antigen is associated with abdominal visceral fat accumulation in female Korean nonsmokers.

BACKGROUND: Carcinoembryonic antigen (CEA) is a tumor marker overexpressed in adenocarcinoma that has proinflammatory properties. Recent studies have reported that CEA is positively associated with carotid atherosclerosis and metabolic syndrome. Because visceral obesity is a known risk factor for cardiometabolic diseases, CEA may also be associated with visceral adiposity. Therefore, we investigated the relationship between serum CEA concentration and visceral obesity in female Korean nonsmokers. METHODS: A total of 270 Korean female nonsmokers were enrolled during their routine health check-ups. Biomarkers of metabolic risk factors were assessed along with body composition by computed tomography. Serum CEA levels were measured by using a chemiluminescence immunoassay analyzer. RESULTS: Serum CEA levels correlated with visceral fat area, fasting glucose, and triglyceride levels after adjusting for age and BMI. The mean visceral fat area increased significantly with the increasing CEA tirtiles. In a step-wise multiple regression analysis, age (ß = 0.26, p<0.01) and visceral fat area (ß = 0.19, p = 0.03) were identified as explanatory variables for serum CEA level. CONCLUSIONS: This study suggested that CEA may be a mediator that links metabolic disturbance and tumorigenesis in visceral obesity. Further studies are required to better understand the clinical and pathophysiological significance of our findings.