Prediction of sarcopenia at different time intervals: an interpretable machine learning analysis of modifiable factors

BMC Geriatrics, Год журнала: 2025, Номер 25(1)

Опубликована: Фев. 27, 2025

This study aims to develop sarcopenia risk prediction models for Chinese older adults at different time intervals and identify compare modifiable factors contributing development. used data from 3,549 participants aged 60 in the China Health Retirement Longitudinal Study (CHARLS). Sarcopenia status was evaluated by AWGS2019 algorithm. Full 2- 4-year risk, considering multifactorial baseline variables, were compared with models. Eight machine learning (ML) algorithms build these models, performance area under receiver operating characteristic curve (AUC-ROC). SHapley Additive exPlanations (SHAP) applied model explanation. The average age of 67.0 years (SD = 6.1), 47.8% being female (1,696 participants). ML achieved moderate performance, eXtreme Gradient Boosting (XGBoost) emerged as best both full 2-year prediction, AUCs 0.804 0.795, respectively (DeLong test, P 0.665). In contrast, Light Machine (LightGBM) performed 0.795 0.769, (P 0.053). SHAP analysis highlighted gender estimated glomerular filtration rate (eGFR) most important predictors Prediction based on can help high sarcopenia. These findings highlight importance prioritizing functional capacity psychosocial determinants prevention strategies.

Язык: Английский

Chest CT-determined sarcopenia is associated with poorer functional outcomes in osteoarthritis patients undergoing total knee arthroplasty: a retrospective cohort study

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Май 25, 2025

Sarcopenia diagnosed via opportunistic computed tomography (CT) is linked to poor postoperative outcomes. This study aimed investigate the association between chest CT-diagnosed sarcopenia and post-total knee arthroplasty (TKA) functional single-center retrospective cohort included 158 osteoarthritis (KOA) patients who underwent primary TKA January 2021 June 2022. was defined using skeletal muscle index values derived from CT images at T12 vertebral level. Sociodemographic, clinical, perioperative data were collected, outcomes assessed Hospital for Special Surgery (HSS) score 6 months postoperatively. Univariate multivariate analyses identified independent risk factors poorer in KOA undergoing TKA. The prevalence of 48.7%. At after TKA, sarcopenic had significantly lower HSS scores (P < 0.05). Advanced age = 0.025), higher body mass 0.027), 0.020), pain 0.001) independently associated with results indicate that CT-determined prevalent among

Язык: Английский

The Prevalence of Malnutrition and Sarcopenia and the Relationship with Inflammation and Anemia Among Community-Dwelling Older Adults: A Preliminary Cross-Sectional Study

Geriatrics, Год журнала: 2024, Номер 9(6), С. 146 - 146

Опубликована: Ноя. 7, 2024

: Older people are more likely to have poor nutrition and low muscle mass, which leads physical performance anemia, resulting in a quality of life risks mobility mortality. Furthermore, malnutrition may, part, raise the level inflammatory biomarkers as well catabolism. Moreover, range indices related systemic inflammation, obtained from routine complete blood count (CBC) tests, been applied inflammation markers. However, these remain insufficiently addressed evidence supporting presence sarcopenia malnutrition. This study aimed explore terms malnutrition, among Thai community-dwelling older people.

Язык: Английский