Deciphering the Effect of Sulfide Derivatives on the Prediction of Nitrite Accumulation in Sulfide-Dosed Partial Nitrification Using Machine Learning

Journal of Environmental Engineering, Год журнала: 2025, Номер 151(7)

Опубликована: Май 1, 2025

Язык: Английский

---

Development and evaluation of interpretable machine learning regressors for predicting femoral neck bone mineral density in elderly men using NHANES data

Biomolecules and Biomedicine, Год журнала: 2024, Номер unknown

Опубликована: Июль 5, 2024

Osteoporotic femoral neck fractures (OFNFs) pose a significant orthopedic challenge in the elderly population, accounting for up to 40% of all osteoporotic and leading considerable health deterioration increased mortality. In addressing critical need early identification osteoporosis through routine screening bone mineral density (FNBMD), this study developed user-friendly prediction model aimed at men aged 50 years older, demographic often overlooked screening. Utilizing data from National Health Nutrition Examination Survey (NHANES), involved outlier detection handling, missing value imputation via K nearest neighbor (KNN) algorithm, normalization encoding. The dataset was split into training test sets with 7:3 ratio, followed by feature least absolute shrinkage selection operator (LASSO) Boruta algorithm. Eight different machine learning algorithms were then employed construct predictive models, their performance evaluated comprehensive metric suite. random forest regressor (RFR) emerged as most effective model, characterized key predictors such age, body mass index (BMI), poverty income ratio (PIR), serum calcium, race, achieving coefficient determination (R²) 0.218 maintaining robustness sensitivity analyses. Notably, excluding race resulted sustained high performance, underscoring model’s adaptability. Interpretations using Shapley additive explanations (SHAP) highlighted influence each on FNBMD. These findings indicate that our effectively aids osteoporosis, potentially reducing incidence OFNFs high-risk population.

Язык: Английский

---

Quantitative expression of LNAPL pollutant concentrations in capillary zone by coupling multiple environmental factors based on random forest algorithm

Journal of Hazardous Materials, Год журнала: 2024, Номер 479, С. 135695 - 135695

Опубликована: Авг. 28, 2024

Язык: Английский

---

Polymorph-Specific Solubility Prediction using Constant Chemical Potential Molecular Dynamics

Опубликована: Март 21, 2024

Molecular Dynamics (MD) simulations offer a robust approach to understanding material properties within system. Solubility is defined as the analytical composition of saturated solution expressed proportion designated solute in solvent, according IUPAC. It critical property compounds and holds significance across numerous fields. Various computational techniques have been explored for determining solubility, including methods based on chemical potential determination, enhanced sampling simulation, direct coexistence lately, machine learning-based shown promise. In this investigation, we aim find solubility values compound through Constant Chemical Potential Dynamics, method rooted simulation. The primary purpose using overcome limitation simulation by maintaining constant sufficiently long time. Urea chosen prototypical system our study, with particular focus three its polymorphs. Our effectively discriminates between polymorphs urea their respective values; polymorph III found highest followed form IV I.

Язык: Английский

---

Polymorph-Specific Solubility Prediction of Urea Using Constant Chemical Potential Molecular Dynamics Simulations

The Journal of Physical Chemistry B, Год журнала: 2024, Номер 128(35), С. 8477 - 8483

Опубликована: Авг. 26, 2024

Molecular dynamics simulations offer a robust approach to understanding the material properties within system. Solubility is defined as analytical composition of saturated solution expressed proportion designated solute in solvent, according IUPAC. It critical property compounds and holds significance across numerous fields. Various computational techniques have been explored for determining solubility, including methods based on chemical potential determination, enhanced sampling simulation, direct coexistence lately, machine learning-based shown promise. In this investigation, we utilized Constant Chemical Potential Dynamics, method rooted predict solubility urea polymorphs aqueous solution. The primary purpose using overcome limitation simulation by maintaining constant sufficiently long time. Urea chosen prototypical system our study, with particular focus three its polymorphs. Our effectively discriminates between their respective values; polymorph III found highest followed forms IV I.

Язык: Английский

---