Objective: This study aimed to develop machine learning-based prediction models to predict masked hypertension and masked uncontrolled hypertension using the clinical characteristics of patients at a single outpatient visit.Methods: Data were derived from two cohorts in Taiwan. The first cohort included 970 hypertensive patients recruited from six medical centers between 2004 and 2005, which were split into a training set (n = 679), a validation set (n = 146), and a test set (n = 145) for model development and internal validation. The second cohort included 416 hypertensive patients recruited from a single medical center between 2012 and 2020, which was used for external validation. We used 33 clinical characteristics as candidate variables to develop models based on logistic regression (LR), random forest (RF), eXtreme Gradient Boosting (XGboost), and artificial neural network (ANN).Results: The four models featured high sensitivity and high negative predictive value (NPV) in internal validation (sensitivity = 0.914–1.000; NPV = 0.853–1.000) and external validation (sensitivity = 0.950–1.000; NPV = 0.875–1.000). The RF, XGboost, and ANN models showed much higher area under the receiver operating characteristic curve (AUC) (0.799–0.851 in internal validation, 0.672–0.837 in external validation) than the LR model. Among the models, the RF model, composed of 6 predictor variables, had the best overall performance in both internal and external validation (AUC = 0.851 and 0.837; sensitivity = 1.000 and 1.000; specificity = 0.609 and 0.580; NPV = 1.000 and 1.000; accuracy = 0.766 and 0.721, respectively).Conclusion: An effective machine learning-based predictive model that requires data from a single clinic visit may help to identify masked hypertension and masked uncontrolled hypertension.

Background and AimsParkinson’s disease (PD) is a worldwide neurodegenerative disease with an increasing global burden, while constipation is an important risk factor for PD. The gastrointestinal tract had been proposed as the origin of PD in Braak’s gut–brain axis hypothesis, and there is increasing evidence indicating that intestinal microbial alteration has a role in the pathogenesis of PD. In this study, we aim to investigate the role of intestinal microbial alteration in the mechanism of constipation-related PD.MethodsWe adapted our data from Hill‐Burns et al., in which 324 participants were enrolled in the study. The 16S rRNA gene sequence data were processed, aligned, and categorized using DADA2. Mediation analysis was used to test and quantify the extent by which the intestinal microbial alteration explains the causal effect of constipation on PD incidence.ResultsWe found 18 bacterial genera and 7 species significantly different between groups of constipated and non-constipated subjects. Among these bacteria, nine genera and four species had a significant mediation effect between constipation and PD. All of them were short-chain fatty acid (SCFA)-producing bacteria that were substantially related to PD. Results from the mediation analysis showed that up to 76.56% of the effect of constipation on PD was mediated through intestinal microbial alteration.ConclusionOur findings support that gut dysbiosis plays a critical role in the pathogenesis of constipation-related PD, mostly through the decreasing of SCFA-producing bacteria, indicating that probiotics with SCFA-producing bacteria may be promising in the prevention and treatment of constipation-related PD.Limitations1) Several potential confounders that should be adjusted were not provided in the original dataset. 2) Our study was conducted based on the assumption of constipation being the etiology of PD; however, constipation and PD may mutually affect each other. 3) Further studies are necessary to explain the remaining 23.44% effect leading to PD by constipation.

Aims The detection of white-coat and white-coat uncontrolled hypertension with out-of-office blood pressure monitoring is time- and resource-consuming. We aim to develop a machine learning-derived prediction model based on the characteristics of patients from a single outpatient visit. Methods and Results Data from two cohorts in Taiwan were used. Cohort one (970 patients) was used for development and internal validation and cohort two (464 patients) was used for external validation. White-coat/white-coat uncontrolled hypertension was defined as an office blood pressure of ≥140/90 mmHg and daytime ambulatory blood pressure of <135/85 mmHg in treatment-naïve or treated individuals. Logistic regression, random forest, extreme gradient boosting, and artificial neural network models were trained using 26 patient parameters. We utilised SHapley Additive exPlanations values to provide explanations for the risk factors. All models achieved great area under the receiver operating characteristic curve, specificity, and negative predictive value in both validations (area under the receiver operating characteristic curve = 0.754–0.891; specificity = 0.682–0.910; negative predictive value = 0.831–0.968). The random forest model was the best performing (area under the receiver operating characteristic curve = 0.884; sensitivity = 0.619; specificity = 0.887; negative predictive value = 0.872; accuracy = 0.819). The five most influential features of the random forest model were office diastolic blood pressure, office systolic blood pressure, current smoker, estimated glomerular filtration rate, and fasting glucose level. Conclusions Our prediction models achieved good performance, underlining the feasibility of applying machine learning models to outpatient populations for the diagnosis of white-coat hypertension and white-coat uncontrolled hypertension. Further validation with a prospective dataset should be considered in the future.