As osteoporosis is a degenerative disease related to postmenopausal aging, early diagnosis is vital. This study used data from the Korea National Health and Nutrition Examination Surveys to predict a patient’s risk of osteoporosis using machine learning algorithms. Data from 1431 postmenopausal women aged 40–69 years were used, including 20 features affecting osteoporosis, chosen by feature importance and recursive feature elimination. Random Forest (RF), AdaBoost, and Gradient Boosting (GBM) machine learning algorithms were each used to train three models: A, checkup features; B, survey features; and C, both checkup and survey features, respectively. Of the three models, Model C generated the best outcomes with an accuracy of 0.832 for RF, 0.849 for AdaBoost, and 0.829 for GBM. Its area under the receiver operating characteristic curve (AUROC) was 0.919 for RF, 0.921 for AdaBoost, and 0.908 for GBM. By utilizing multiple feature selection methods, the ensemble models of this study achieved excellent results with an AUROC score of 0.921 with AdaBoost, which is 0.1–0.2 higher than those of the best performing models from recent studies. Our model can be further improved as a practical medical tool for the early diagnosis of osteoporosis after menopause.
1 January 2021 in Archives of Gerontology and Geriatrics
Diabetic kidney disease (DKD) is one of the most common forms of chronic kidney disease. Its pathogenic mechanism is complex, and it can affect entire structures of the kidney. However, conventional approaches to early stage DKD have focused on changes to the glomerulus. Current standard screening tools for DKD, albuminuria, and estimated glomerular filtration rate are insufficient to reflect early tubular injury. Therefore, many tubular biomarkers have been suggested. Non-albumin proteinuria (NAP) contains a wide range of tubular biomarkers and is convenient to measure. We reviewed the clinical meanings of NAP and its significance as a marker for early stage DKD.
Background: Obesity and metabolic syndrome (MetS) are increasing rapidly worldwide, but metabolically healthy population exists. Thus, this study evaluated the proportion of obesity and MetS and investigated clinical and body compositional factors associated with MetS in obese Koreans. Methods: This cross-sectional observational study enrolled subjects from the Korea National Health and Nutrition Examination Survey (n = 15,408). Obesity was defined by body mass index (BMI) cutoff points of 25 and 27.5 kg/m2. Clinical and laboratory parameters were compared by MetS presence and factors associated with insulin resistance or MetS in obese subjects were evaluated by multivariate or logistic regression analyses. Results: The proportion of obesity was 29.7% at a BMI cutoff point of 25 kg/m2 and 10.4% at 27.5 kg/m2. The proportion of MetS was greater in obese population and increased with higher obesity cutoff point, where men were more prevalent than women. The subjects with MetS showed higher age, glycated hemoglobin, insulin resistance, liver enzymes, and ferritin compared with metabolically healthy population in both obese genders. Factors associated with insulin level or presence of MetS were high trunk fat percentage, trunk-to-limb fat ratio, and trunk-to-leg fat ratio, but low limb fat percentage and leg fat percentage. Among body composition indices, trunk-to-limb fat ratio and trunk-to-leg fat ratio showed significantly high odds ratios for MetS in both obese genders regardless of BMI cutoff points. Conclusions: MetS in obese population concurrently increased with higher obesity cutoff point. Higher age, insulin resistance, liver enzymes, ferritin, and higher proportion of truncal fat mass-to-limb or leg fat mass significantly contribute to MetS, where trunk-to-limb fat ratio and trunk-to-leg fat ratio could be helpful in identifying and preventing MetS in the obese Korean population.