Pingdingshan University

Fetal DNA concentration (FF) is a critical indicator affecting the accuracy and stability of non-invasive prenatal testing (NIPT). An excessively low FF may lead to test failure or false negatives. To address the lack of quantitative prediction for the earliest safe testing time at which clinical thresholds (FF ≥4%) are met, this study constructed multivariate linear regression and Logistic regression models based on publicly available sample data. The models analyzed the influence of gestational age and maternal body mass index (BMI) on FF levels and the probability of meeting the threshold, and introduced a gestational age quadratic term to establish a nonlinear model. The results demonstrated that BMI was a significant negative predictor of both FF and the probability of meeting the threshold. Further, the study derived the earliest gestational age at which the probability of meeting the threshold ≥95% for different BMI stratifications and validated model stability using the Bootstrap method. This research provides a quantitative basis for individualized NIPT testing timing.

ABSTRACT The evolution of smart grid and the deepening of the power market have significantly enhanced the key position of precise electricity price estimation in industry decision‐making and the optimal resource allocation. It is necessary to implement short‐term electricity price forecast in the power market that accurately characterizes the dynamic change laws of electricity prices in a market environment with significant high volatility and nonlinear characteristics. To address this demand, the paper proposes an Attention‐based long short‐term memory model called ATT‐LSTM, which intends to improve the accuracy and interpretability of electricity price forecast, and provide support for power market operation analysis and trading decision making. In the model, multi‐dimensional feature engineering and data preprocessing are first performed, including meteorological data imputation, outlier removal, and time series sliding window construction. An attention layer is introduced to dynamically allocate importance weights to different time steps and features, allowing the model to adaptively focus on key influencing variables. This model is particularly suitable for electricity price forecast scenarios with significant seasonality and short‐term disturbances. Experiments are based on measured data from the Australian National Electricity Market, with comparisons conducted using multiple benchmark models, that is, LSTM, BPNN, and ATT‐LSTM. The results show that the ATT‐LSTM model achieves the optimal performance in terms of multiple metrics (e.g., MAE and RMSE) and prove that introducing the attention mechanism into the LSTM can effectively improve the accuracy and transparency of electricity price forecast.

Reforming 3d-metal-based visible light catalytic platforms is desirable yet challenging for the selective cleavage of the C–C bond in lignin to value-added biochemicals. Herein, we provide a cost-effective iron-catalyzed photochemical strategy for the selective conversion of lignin to benzaldehyde under simulated natural conditions. Furthermore, the product distribution can be rationally regulated by changing the counteranion of Fe3+. With Fe(NO3)3·9H2O, a 100% conversion of diphenyl ethanol (lignin model) was achieved, affording a 186.0 mol % yield and 93.0% selectivity for benzaldehyde via the Cα–Cβ bond cleavage. By contrast, FeCl3·6H2O predominantly favored the Cα–OH oxidation to form diphenylethanone (72.4% selectivity). The results of the mechanistic study and density functional theory (DFT) calculation unveil that benzaldehyde formation proceeds via β-scission of an FeIII alkoxide intermediate through photodriven ligand-to-metal charge transfer (LMCT), wherein the nitrate counteranion serves as an internal oxidant in the iron nitrate catalytic system. Conversely, FeCl3 generates chlorine radicals via homolytic cleavage, resulting in the hydrogen atom abstraction at Cα–OH which consequently inhibits the breakage of the Cα–Cβ bond. Notably, the Fe(NO3)3 catalytic system also enables efficient C–C bond cleavage in realistic lignin (121.3 mg g–1 yield of monophenols), as evidenced by 2D HSQC NMR and FT-IR. Therefore, the findings in this work advance solar-driven lignin valorization and, more importantly, offer deep insights into the recently reemerging photochemistry of FeIII salts.