Dataset Information

Development and Validation of a Deep Learning Based Diabetes Prediction System Using a Nationwide Population-Based Cohort.

ABSTRACT:

Background

Previously developed prediction models for type 2 diabetes mellitus (T2DM) have limited performance. We developed a deep learning (DL) based model using a cohort representative of the Korean population.

Methods

This study was conducted on the basis of the National Health Insurance Service-Health Screening (NHIS-HEALS) cohort of Korea. Overall, 335,302 subjects without T2DM at baseline were included. We developed the model based on 80% of the subjects, and verified the power in the remainder. Predictive models for T2DM were constructed using the recurrent neural network long short-term memory (RNN-LSTM) network and the Cox longitudinal summary model. The performance of both models over a 10-year period was compared using a time dependent area under the curve.

Results

During a mean follow-up of 10.4±1.7 years, the mean frequency of periodic health check-ups was 2.9±1.0 per subject. During the observation period, T2DM was newly observed in 8.7% of the subjects. The annual performance of the model created using the RNN-LSTM network was superior to that of the Cox model, and the risk factors for T2DM, derived using the two models were similar; however, certain results differed.

Conclusion

The DL-based T2DM prediction model, constructed using a cohort representative of the population, performs better than the conventional model. After pilot tests, this model will be provided to all Korean national health screening recipients in the future.

SUBMITTER: Rhee SY

PROVIDER: S-EPMC8369223 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundGestational diabetes (GDM) is prevalent and benefits from timely and effective treatment, given the short window to impact glycemic control. Clinicians face major barriers to choosing effectively among treatment modalities [medical nutrition therapy (MNT) with or without pharmacologic treatment (antidiabetic oral agents and/or insulin)]. We investigated whether clinical data at varied stages of pregnancy can predict GDM treatment modality.MethodsAmong a population-based cohort of 30,474 pregnancies with GDM delivered at Kaiser Permanente Northern California in 2007-2017, we selected those in 2007-2016 as the discovery set and 2017 as the temporal/future validation set. Potential predictors were extracted from electronic health records at different timepoints (levels 1-4): (1) 1-year preconception to the last menstrual period, (2) the last menstrual period to GDM diagnosis, (3) at GDM diagnosis, and (4) 1 week after GDM diagnosis. We compared transparent and ensemble machine learning prediction methods, including least absolute shrinkage and selection operator (LASSO) regression and super learner, containing classification and regression tree, LASSO regression, random forest, and extreme gradient boosting algorithms, to predict risks for pharmacologic treatment beyond MNT.ResultsThe super learner using levels 1-4 predictors had higher predictability [tenfold cross-validated C-statistic in discovery/validation set: 0.934 (95% CI: 0.931-0.936)/0.815 (0.800-0.829)], compared to levels 1, 1-2, and 1-3 (discovery/validation set C-statistic: 0.683-0.869/0.634-0.754). A simpler, more interpretable model, including timing of GDM diagnosis, diagnostic fasting glucose value, and the status and frequency of glycemic control at fasting during one-week post diagnosis, was developed using tenfold cross-validated logistic regression based on super learner-selected predictors. This model compared to the super learner had only a modest reduction in predictability [discovery/validation set C-statistic: 0.825 (0.820-0.830)/0.798 (95% CI: 0.783-0.813)].ConclusionsClinical data demonstrated reasonably high predictability for GDM treatment modality at the time of GDM diagnosis and high predictability at 1-week post GDM diagnosis. These population-based, clinically oriented models may support algorithm-based risk-stratification for treatment modality, inform timely treatment, and catalyze more effective management of GDM.

Project description:To explore the risk factors of developing chronic pancreatitis (CP) in patients with acute pancreatitis (AP) and develop a prediction score for CP. Using the National Health Insurance Research Database in Taiwan, we obtained large, population-based data of 5971 eligible patients diagnosed with AP from 2000 to 2013. After excluding patients with obstructive pancreatitis and biliary pancreatitis and those with a follow-up period of less than 1 year, we conducted a multivariate analysis using the data of 3739 patients to identify the risk factors of CP and subsequently develop a scoring system that could predict the development of CP in patients with AP. In addition, we validated the scoring system using a validation cohort. Among the study subjects, 142 patients (12.98%) developed CP among patients with RAP. On the other hand, only 32 patients (1.21%) developed CP among patients with only one episode of AP. The multivariate analysis revealed that the presence of recurrent AP (RAP), alcoholism, smoking habit, and age of onset of < 55 years were the four important risk factors for CP. We developed a scoring system (risk score 1 and risk score 2) from the derivation cohort by classifying the patients into low-risk, moderate-risk, and high-risk categories based on similar magnitudes of hazard and validated the performance using another validation cohort. Using the prediction score model, the area under the curve (AUC) [95% confidence interval (CI)] in predicting the 5-year CP incidence in risk score 1 (without the number of AP episodes) was 0.83 (0.79, 0.87), whereas the AUC (95%CI) in risk score 2 (including the number of AP episodes) was 0.84 (0.80, 0.88). This result demonstrated that the risk score 2 has somewhat better prediction performance than risk score 1. However, both of them had similar performance between the derivation and validation cohorts. In the study,we identified the risk factors of CP and developed a prediction score model for CP.