The University of Osaka · Health Professions
Professor Hiroe Seto's research lab specializes in health data science and machine learning applications for preventive medicine, with a focus on developing and validating reliable risk prediction models for chronic diseases such as diabetes and metabolic syndrome. The lab emphasizes rigorous calibration assessment in predictive modeling, particularly through innovative methods like the variable-based probabilistic calibration plot (VPC-Plot) to improve model reliability for clinically important variables. Their work leverages large-scale population health data, such as the Kokuho-database in Osaka, Japan, to explore seasonal and demographic variations in metabolic health markers. The lab is at the forefront of advancing statistical methodology in machine learning for public health, ensuring models are not only accurate but also clinically trustworthy.
Figures are computed from collected data and may differ slightly.
We sought to verify the reliability of machine learning (ML) in developing diabetes prediction models by utilizing big data. To this end, we compared the reliability of gradient boosting decision tree (GBDT) and logistic regression (LR) models using data obtained from the Kokuho-database of the Osaka prefecture, Japan. To develop the models, we focused on 16 predictors from health checkup data from April 2013 to December 2014. A total of 277,651 eligible participants were studied. The prediction
This finding, complex seasonal variations of MetS prevalence, WC, TG, and FPG, could not be derived from previous studies using just the mean values in spring, summer, autumn and winter or the cosinor analysis. More attention should be paid to factors affecting seasonal variations of central obesity, dyslipidemia and insulin resistance.
In developing risk prediction models for specific diseases, it is essential to evaluate the calibration performance of the prediction model. Various methods have been proposed to assess the calibration of prediction models, but it has been pointed out that conventional methods based on the predicted probability of the model are insufficient to detect miscalibration. Another problem is that a method for evaluating calibration for continuous variables of interest has not yet been established. We t
<title>Abstract</title> We sought to verify the reliability of machine learning (ML) in developing diabetes prediction models by utilizing big data. To this end, we compared the reliability of gradient boosting decision tree (GBDT) and logistic regression (LR) models using data obtained from the Kokuho-database of the Osaka prefecture, Japan. To develop the models, we focused on 16 predictors from health checkup data from April 2013 to December 2014. A total of 277,651 eligible participants were
<title>Abstract</title> In developing risk prediction models for specific diseases, it is essential to evaluate the calibration performance of the prediction model. Various methods have been proposed to assess the calibra- tion of prediction models, but it has been pointed out that conventional methods based on the predicted probability of the model are insufficient to detect miscalibration. Another problem is the inability to assess calibration for variables of interest, such as covariate of hi
Open papers in the app to read, cite, and organize with AI.
