Leveraging machine learning for precision medicine: a predictive model for cognitive impairment in cholestasis patients

BMC Gastroenterology, Journal Year: 2025, Volume and Issue: 25(1)

Published: March 18, 2025

Cholestasis, characterized by impaired bile flow, impacts cognitive function through systemic mechanisms, including inflammation and metabolic dysregulation. Despite its significance, targeted predictive models for impairment in cholestasis remain underexplored. This study addresses this gap developing a machine learning-based model tailored to population. Clinical biochemical data from Qingyang People's Hospital (2021–2023) were used train validate predicting (MoCA ≤ 17). Recursive feature elimination identified critical predictors, while LightGBM other learning evaluated. SHAP analysis enhanced interpretability, clinical utility was assessed decision curve (DCA). outperformed with an AUC of 0.7955 on the testing dataset. Age, plasma D-dimer, albumin key predictors. revealed non-linear interactions among features, demonstrating model's alignment. DCA confirmed improving patient stratification. The developed LightGBM-based effectively predicts patients, providing actionable insights early intervention. Integrating tool into workflows can enhance precision medicine improve outcomes high-risk

Language: Английский

Genetic strategies for enhancing litter size and birth weight uniformity in piglets

Frontiers in Veterinary Science, Journal Year: 2025, Volume and Issue: 12

Published: March 24, 2025

This study aimed to estimate the genetic parameters and develop selection indices for litter size birth weight uniformity in piglets. These traits are crucial improving productivity profitability of swine production. Data were collected from 9,969 litters 4,465 sows 106,463 piglets various breeds a farm Thailand. The analyzed included total number born (TNB), alive (NBA), (LBW), mean weight, individual weight. assessed piglet difference between maximum minimum values (range), interquartile range (IQRBW), variance (VBW), standard deviation (SDBW), coefficient variation (CVBW). Variance components estimated using multiple-trait animal model average information-restricted likelihood method. appropriate index (I) was determined based on heritability, correlations traits, economic significance traits. results revealed that (TNB NBA) (Range, IQRBW, VBW, SDBW, CVBW) exhibited low heritability ( p < 0.1), suggesting environmental factors have substantial influence. In contrast, showed moderate (approximately 0.2). Negative observed, indicating increasing might reduce uniformity, potentially affecting survival rate. A combining NBA, LBW, CVBW constructed optimize process uniformity. conclusion, improvement programs should prioritize enhance commercial pig farms. findings can assist breeders developing more effective strategies, ultimately resulting larger, uniform improved overall efficiency.

Language: Английский