Dataset Information

Prediction of Incident Diabetes in the Jackson Heart Study Using High-Dimensional Machine Learning.

ABSTRACT: Statistical models to predict incident diabetes are often based on limited variables. Here we pursued two main goals: 1) investigate the relative performance of a machine learning method such as Random Forests (RF) for detecting incident diabetes in a high-dimensional setting defined by a large set of observational data, and 2) uncover potential predictors of diabetes. The Jackson Heart Study collected data at baseline and in two follow-up visits from 5,301 African Americans. We excluded those with baseline diabetes and no follow-up, leaving 3,633 individuals for analyses. Over a mean 8-year follow-up, 584 participants developed diabetes. The full RF model evaluated 93 variables including demographic, anthropometric, blood biomarker, medical history, and echocardiogram data. We also used RF metrics of variable importance to rank variables according to their contribution to diabetes prediction. We implemented other models based on logistic regression and RF where features were preselected. The RF full model performance was similar (AUC = 0.82) to those more parsimonious models. The top-ranked variables according to RF included hemoglobin A1C, fasting plasma glucose, waist circumference, adiponectin, c-reactive protein, triglycerides, leptin, left ventricular mass, high-density lipoprotein cholesterol, and aldosterone. This work shows the potential of RF for incident diabetes prediction while dealing with high-dimensional data.

SUBMITTER: Casanova R

PROVIDER: S-EPMC5058485 | biostudies-literature | 2016

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of Incident Diabetes in the Jackson Heart Study Using High-Dimensional Machine Learning.

Casanova Ramon R Saldana Santiago S Simpson Sean L SL Lacy Mary E ME Subauste Angela R AR Blackshear Chad C Wagenknecht Lynne L Bertoni Alain G AG

PloS one 20161011 10

Statistical models to predict incident diabetes are often based on limited variables. Here we pursued two main goals: 1) investigate the relative performance of a machine learning method such as Random Forests (RF) for detecting incident diabetes in a high-dimensional setting defined by a large set of observational data, and 2) uncover potential predictors of diabetes. The Jackson Heart Study collected data at baseline and in two follow-up visits from 5,301 African Americans. We excluded those w ...[more]

PMID: 27727289

Similar Datasets

Project description:ObjectivesEvaluation of the feasibility of using cardiovascular magnetic resonance (CMR) radiomics in the prediction of incident atrial fibrillation (AF), heart failure (HF), myocardial infarction (MI), and stroke using machine learning techniques.MethodsWe identified participants from the UK Biobank who experienced incident AF, HF, MI, or stroke during the continuous longitudinal follow-up. The CMR indices and the vascular risk factors (VRFs) as well as the CMR images were obtained for each participant. Three-segmented regions of interest (ROIs) were computed: right ventricle cavity, left ventricle (LV) cavity, and LV myocardium in end-systole and end-diastole phases. Radiomics features were extracted from the 3D volumes of the ROIs. Seven integrative models were built for each incident cardiovascular disease (CVD) as an outcome. Each model was built with VRF, CMR indices, and radiomics features and a combination of them. Support vector machine was used for classification. To assess the model performance, the accuracy, sensitivity, specificity, and AUC were reported.ResultsAF prediction model using the VRF+CMR+Rad model (accuracy: 0.71, AUC 0.76) obtained the best result. However, the AUC was similar to the VRF+Rad model. HF showed the most significant improvement with the inclusion of CMR metrics (VRF+CMR+Rad: 0.79, AUC 0.84). Moreover, adding only the radiomics features to the VRF reached an almost similarly good performance (VRF+Rad: accuracy 0.77, AUC 0.83). Prediction models looking into incident MI and stroke reached slightly smaller improvement.ConclusionsRadiomics features may provide incremental predictive value over VRF and CMR indices in the prediction of incident CVDs.Key points• Prediction of incident atrial fibrillation, heart failure, stroke, and myocardial infarction using machine learning techniques. • CMR radiomics, vascular risk factors, and standard CMR indices will be considered in the machine learning models. • The experiments show that radiomics features can provide incremental predictive value over VRF and CMR indices in the prediction of incident cardiovascular diseases.

Project description:Background The prognostic value of anthropometric, adipokine, and computed tomography measures of adiposity to predict diabetes in Black, specifically by normoglycemia versus prediabetes, remains incompletely understood. Methods and Results Among Black participants without diabetes in the JHS (Jackson Heart Study), waist circumference [WC], body mass index, adiponectin, leptin, and leptin:adiponectin ratio were standardized in sample 1 (2422 participants at baseline [2000-2004]) and WC, body mass index, visceral adipose tissue (VAT), subcutaneous adipose tissue, and liver attenuation in 1537 participants at examination 2 (2005-2008) (sample 2). Hazard ratios (HRs) for diabetes were estimated using interval-censored Cox modeling adjusting for traditional risk factors and validated with the C index. Over 5 years, 300 and 122 incident diabetes cases occurred in sample 1 and sample 2, respectively. In sample 1 and sample 2, a 1-SD higher log-leptin:adiponectin ratio and VAT had the strongest associations (HR, 1.95 [95% CI, 1.67-2.27] and 1.76 [95% CI, 1.52-2.04]) and discriminatory power (C index 0.68 [95% CI, 0.64-0.71] and C index 0.67 [95% CI, 0.61-0.74]) with diabetes. The normoglycemic compared with the prediabetes group had a 1.3 to 1.9 times greater magnitude of associations with diabetes for WC, liver attenuation, and VAT (P interaction <0.10). In sample 2, C indices for WC (HR, 0.84; 95% CI, 0.73-0.95), VAT (HR, 0.91; 95% CI, 0.85-0.98), and liver attenuation (HR, 0.90; 95% CI, 0.77-1.00) were greater than HbA1c (HR, 0.74; 95% CI, 0.57-0.90) in normoglycemia, whereas HbA1c was best in prediabetes (HR, 0.72; 95% CI, 0.66-0.78). Conclusions Overall, among Black adults, multiple measures of adiposity were associated with incident diabetes with modest predictive ability. In Black patients with normoglycemia, WC, liver attenuation, and VAT may appropriately identify those at high risk for diabetes, whereas HbA1c was the best predictor in individuals with prediabetes.

Dataset Information

Prediction of Incident Diabetes in the Jackson Heart Study Using High-Dimensional Machine Learning.

Publications

Prediction of Incident Diabetes in the Jackson Heart Study Using High-Dimensional Machine Learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets