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 models were developed using a light gradient boosting machine (LightGBM), which is an effective GBDT implementation algorithm, and LR. Their reliabilities were measured based on expected calibration error (ECE), negative log-likelihood (Logloss), and reliability diagrams. Similarly, their classification accuracies were measured in the area under the curve (AUC). We further analyzed their reliabilities while changing the sample size for training. Among the 277,651 participants, 15,900 (7978 males and 7922 females) were newly diagnosed with diabetes within 3 years. LightGBM (LR) achieved an ECE of 0.0018 ± 0.00033 (0.0048 ± 0.00058), a Logloss of 0.167 ± 0.00062 (0.172 ± 0.00090), and an AUC of 0.844 ± 0.0025 (0.826 ± 0.0035). From sample size analysis, the reliability of LightGBM became higher than LR when the sample size increased more than $$10^4$$
10
4
. Thus, we confirmed that GBDT provides a more reliable model than that of LR in the development of diabetes prediction models using big data. ML could potentially produce a highly reliable diabetes prediction model, a helpful tool for improving lifestyle and preventing diabetes.
It is generally believed that electromagnetic noise originates from the coupling of electric signals in a circuit with electric signals in surrounding materials in the environment. However, the noise phenomenon had not been quantified until now. In order to study the phenomenon of noise, we considered a standard circuit (two transmission lines), to which an additional transmission line was introduced in order to explicitly take into account the effect of conductors in the environment. We performed calculations using a newly developed multiconductor transmission-line theory for the resulting three-line circuit in order to determine the magnitude of the coupling between the circuit and the conductors in the environment under various conditions. We observed ringing and resonance phenomena in the common mode, which influenced the performance of the normal mode as electromagnetic noise. Our findings were confirmed by recent experiments in which conductor lines were arranged in various ways using a printed circuit board (PCB). The ordinary usage of electricity in the standard electric circuit was found to be worst in exciting the common mode noise.
We calculate electromagnetic phenomena in the multi-layer plane circuit starting from the Maxwell equations. We present a numerical method of potential and current density in two-dimensional conductors, where their time developments are treated as phenomena of wave propagation. We treat the plane conductors by dividing them into small finite-volume elements, similar to the case of the partial element equivalent circuit method, and the transport equations are then solved by the finite-difference time-domain method. Furthermore, we develop a calculation method for the boundary in a multi-layer plane by applying the method we have used in multi-transmission lines. We formulate the boundary conditions of a multi-layer plane coupled with lumped-parameter circuits and introduce an algorithm to reduce calculation costs that are largely associated with the two-dimensional extension from the multi-transmission-line case. We perform calculations of the wave propagation of potential, current density, and charge density in the time domain for a simple plane circuit. These calculations are presented as supplementary materials of the present paper.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.