Machine Learning Models for Data-Driven Prediction of Diabetes by Lifestyle Type.

The prevalence of diabetes has been increasing in recent years, and previous research has found that machine-learning models are good diabetes prediction tools. The purpose of this study was to compare the efficacy of five different machine-learning models for diabetes prediction using lifestyle data from the National Health and Nutrition Examination Survey (NHANES) database. The 1999-2020 NHANES database yielded data on 17,833 individuals data based on demographic characteristics and lifestyle-related variables. To screen training data for machine models, the Akaike Information Criterion (AIC) forward propagation algorithm was utilized. For predicting diabetes, five machine-learning models (CATBoost, XGBoost, Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM)) were developed. Model performance was evaluated using accuracy, sensitivity, specificity, precision, F1 score, and receiver operating characteristic (ROC) curve. Among the five machine-learning models, the dietary intake levels of energy, carbohydrate, and fat, contributed the most to the prediction of diabetes patients. In terms of model performance, CATBoost ranks higher than RF, LG, XGBoost, and SVM. The best-performing machine-learning model among the five is CATBoost, which achieves an accuracy of 82.1% and an AUC of 0.83. Machine-learning models based on NHANES data can assist medical institutions in identifying diabetes patients.

Prevalence of psychological disorders, sleep disturbance and stressful life events and their relationships with disease parameters in Chinese patients with ankylosing spondylitis.

Our aim was to investigate the prevalence of psychological disorders, sleep disturbance, and stressful life events in Chinese patients with ankylosing spondylitis (AS) and healthy controls, to assess the correlation between psychological and disease-related variables, and finally to detect powerful factors in predicting anxiety and depression. AS patients diagnosed with the modified New York criteria and healthy controls were enrolled from China. Participants completed a set of questionnaires, including demographic and disease parameters, Zung self-rating anxiety scale (SAS), Zung self-rating depression scale (SDS), the Pittsburgh Sleep Quality Index questionnaire (PSQI), and the Social Readjustment Rating Scale (SRRS). The relationship between psychological and other variables was explored. Stepwise multiple regression was used to determine the contributors to each disorder. Of all the 2772 AS patients, 79.1% were male. Mean age was 28.99 ± 8.87 years. Prevalence of anxiety, depression, and sleep disturbance was 31.6% (95% CI, 29.9, to 33.4), 59.3% (95% CI, 57.5, to 61.2), and 31.0% (95% CI, 29.3, to 36.7), respectively. 35.3% had stimulus of psychological and social elements (SPSE). Compared with healthy controls, AS patients had more severe psychological disorders, sleep disturbance, and stressful life events (P < 0.01). SDS, overall pain, BASFI, and sleep disturbance were significant contributors of the SAS scores (P < 0.03). SAS, less years of education, and sleep duration were significant contributors of SDS (P < 0.01). AS patients had more anxiety, depression, stressful life events, and sleep disturbance than healthy controls. Pain, functional limitation, sleep disturbance, and education were major contributors to psychological disorders.

The relationship between disease activity measured by the BASDAI and psychological status, stressful life events, and sleep quality in ankylosing spondylitis.

Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) is a standard instrument regularly used to assess disease activity of patients with ankylosing spondylitis (AS). However, the well-being of a patient is also affected by impairment of function as well as psychological status and other factors. The objective of this study was to evaluate if psychological status, stressful life events, and sleep quality contribute significantly to BASDAI. Six hundred eighty-three AS patients satisfying the Modified New York Criteria for AS were recruited from the rheumatology clinics of seven hospitals in China. Patients with other concomitant disorders were excluded. Participants were requested to complete a set of clinical examinations and the following questionnaires: Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Bath Ankylosing Spondylitis Functional Index (BASFI), Zung Self-Rating Anxiety Scale (SAS), Zung Self-Rating Depression Scale (SDS), Pittsburgh Sleep Quality Index Questionnaire (PSQI), Health Assessment Questionnaire for Spondyloarthropathies (HAQ-S), and Social Readjustment Rating Scale (SRRS). Spearman correlation analysis showed that BASDAI was highly associated with degree and duration of morning stiffness, overall pain, nocturnal back pain, overall back pain, anxiety, and BASFI (all P < 0.001), but were not associated with education, HAQ-S, and sleep medication in PSQI (P > 0.05). Multiple stepwise regression analysis indicated that overall pain was the maximal statistical contribution in predicting disease activity (standardized coefficient, 0.335). In hierarchic multiple regression analysis, psychological variables added an only additional 2.7% to the overall R(2) beyond that accounted for by demographic and medical variables, resulting in a final R(2) of 53.5%. Although BASDAI is a very good measurement of pain and stiffness and to a certain extent effect of functional impairment in AS, it barely takes into account psychological status, stress life events, and sleep quality These factors should be evaluated by other modalities.