Comparing the accuracy of four machine learning models in predicting type 2 diabetes onset within the Chinese population: a retrospective study.

OBJECTIVE: To evaluate the effectiveness of machine learning (ML) models in predicting 5-year type 2 diabetes mellitus (T2DM) risk within the Chinese population by retrospectively analyzing annual health checkup records. METHODS: We included 46,247 patients (32,372 and 13,875 in training and validation sets, respectively) from a national health checkup center database. Univariate and multivariate Cox analyses were performed to identify factors influencing T2DM risk. Extreme Gradient Boosting (XGBoost), support vector machine (SVM), logistic regression (LR), and random forest (RF) models were trained to predict 5-year T2DM risk. Model performances were analyzed using receiver operating characteristic (ROC) curves for discrimination and calibration plots for prediction accuracy. RESULTS: Key variables included fasting plasma glucose, age, and sedentary time. The LR model showed good accuracy with respective areas under the ROC (AUCs) of 0.914 and 0.913 in training and validation sets; the RF model exhibited favorable AUCs of 0.998 and 0.838. In calibration analysis, the LR model displayed good fit for low-risk patients; the RF model exhibited satisfactory fit for low- and high-risk patients. CONCLUSIONS: LR and RF models can effectively predict T2DM risk in the Chinese population. These models may help identify high-risk patients and guide interventions to prevent complications and disabilities.

An adaptive kernel dictionary-based low-rank representation method for subspace clustering.

Low-rank representation (LRR) is a classic subspace clustering (SC) algorithm, and many LRR-based methods have been proposed. Generally, LRR-based methods use denoized data as dictionaries for data reconstruction purpose. However, the dictionaries used in LRR-based algorithms are fixed, leading to poor clustering performance. In addition, most of these methods assume that the input data are linearly correlated. However, in practice, data are mostly nonlinearly correlated. To address these problems, we propose a novel adaptive kernel dictionary-based LRR (AKDLRR) method for SC. Specifically, to explore nonlinear information, the given data are mapped to the Hilbert space via the kernel technique. The dictionary in AKDLRR is not fixed; it adaptively learns from the data in the kernel space, making AKDLRR robust to noise and yielding good clustering performance. To solve the AKDLRR model, an efficient procedure including an alternative optimization strategy is proposed. In addition, a theoretical analysis of the convergence performance of AKDLRR is presented, which reveals that AKDLRR can converge in at most three iterations under certain conditions. The experimental results show that AKDLRR can achieve the best clustering performance and has excellent speed in comparison with other algorithms.

Robust Least Squares Regression for Subspace Clustering: A Multi-View Clustering Perspective.

Recently, with the assumption that samples can be reconstructed by themselves, subspace clustering (SC) methods have achieved great success. Generally, SC methods contain some parameters to be tuned, and different affinity matrices can obtain with different parameter values. In this paper, for the first time, we study a method for fusing these different affinity matrices to promote clustering performance and provide the corresponding solution from a multi-view clustering (MVC) perspective. That is, we argue that the different affinity matrices are consistent and complementary, which is similar to the fundamental assumption of MVC methods. Based on this observation, in this paper, we use least squares regression (LSR), which is a typical SC method, as an example since it can be efficiently optimized and has shown good clustering performance and we propose a novel robust least squares regression method from an MVC perspective (RLSR/MVCP). Specifically, we first utilize LSR with different parameter values to obtain different affinity matrices. Then, to fully explore the information contained in these different affinity matrices and to remove noise, we further fuse these affinity matrices into a tensor, which is constrained by the tensor low-rank constraint, i.e., the tensor nuclear norm (TNN). The two steps are combined into a framework that is solved by the augmented Lagrange multiplier (ALM) method. The experimental results on several datasets indicate that RLSR/MVCP has very encouraging clustering performance and is superior to state-of-the-art SC methods.

NiFeMn-Layered Double Hydroxides Linked by Graphene as High-Performance Electrocatalysts for Oxygen Evolution Reaction.

Currently, precious metal group materials are known as the efficient and widely used oxygen evolution reaction (OER) and hydrogen evolution reaction (HER) catalysts. The exorbitant prices and scarcity of the precious metals have stimulated scale exploration of alternative non-precious metal catalysts with low-cost and high performance. Layered double hydroxides (LDHs) are a promising precursor to prepare cost-effective and high-performance catalysts because they possess abundant micropores and nitrogen self-doping after pyrolysis, which can accelerate the electron transfer and serve as active sites for efficient OER. Herein, we developed a new highly active NiFeMn-layered double hydroxide (NFM LDH) based electrocatalyst for OER. Through building NFM hydroxide/oxyhydroxide heterojunction and incorporation of conductive graphene, the prepared NFM LDH-based electrocatalyst delivers a low overpotential of 338 mV at current density of 10 mA cm-2 with a small Tafel slope of 67 mV dec-1, which are superior to those of commercial RuO2 catalyst for OER. The LDH/OOH heterojunction involves strong interfacial coupling, which modulates the local electronic environment and boosts the kinetics of charge transfer. In addition, the high valence Fe3+ and Mn3+ species formed after NaOH treatment provide more active sites and promote the Ni2+ to higher oxidation states during the O2 evolution. Moreover, graphene contributes a lot to the reduction of charge transfer resistance. The combining effects have greatly enhanced the catalytic ability for OER, demonstrating that the synthesized NFM LDH/OOH heterojunction with graphene linkage can be practically applied as a high-performance electrocatalyst for oxygen production via water splitting.

The predictive ability of plasma ESR1 mutations for the efficacy of endocrine therapy in hormone-receptor-positive advanced breast cancer.

PURPOSE: The predictive ability of plasma ESR1 mutations for outcomes among patients with advanced breast cancer undergoing endocrine therapy (ET) remains disputable. We performed a comprehensive meta-analysis of published studies to clarify the impact of plasma ESR1 mutations on clinical outcomes for patients after subsequent ET. MATERIALS AND METHODS: An electronic search was performed to identify eligible studies. Studies analyzing progression-free survival (PFS) and/or overall survival (OS) according to plasma ESR1 mutation status after subsequent ET were included. HRs were calculated using a fixed- or random-effects model according to heterogeneity. Pooled HRs and 95% CIs were used to estimate the effects. RESULTS: Six studies including 705 patients with advanced breast cancer met the inclusion criteria. The impact of plasma ESR1 mutations on PFS and OS after subsequent ET was reported in six studies (seven groups) and two studies, respectively. Meta-analysis results showed that the pooled HR for ESR1 mutations was 1.70 (95% CI, 1.05-2.74; P=0.03) for OS, which was statistically significant for predicting poor survival, and 1.56 (95% CI, 1.13-2.14; P=0.006) for PFS; however, Begg's and Egger's test results identified the presence of bias. The trim-and-fill method was used, and after incorporation of the imputed studies, the HR was 1.16 (95% CI, 0.88-1.53, P=0.30) for PFS, which indicates that plasma ESR1 mutation had no effect on PFS after subsequent ET. Subgroup analysis suggested that plasma ESR1 mutations were correlated with shorter PFS (HR, 1.98; 95% CI, 1.12-3.51; P=0.02) in patients subsequently treated with aromatase inhibitors (AIs), whereas no association with PFS was observed for patients subsequently treated with non-AI ET (HR, 1.08; 95% CI, 0.85-1.38; P=0.54) or fulvestrant (HR, 1.03; 95% CI, 0.79-1.34; P=0.83). CONCLUSION: The current meta-analysis demonstrates that plasma ESR1 mutation status is not a predictor of ET efficacy for all drugs without distinction in patients with hormone-receptor-positive advanced breast cancer. ESR1 mutation predicted a poor response to AIs, whereas it was not predictive of non-AI ET efficacy, especially for fulvestrant.