fig1

Triglyceride-glucose waist-to-weight index and bone microarchitecture in type 2 diabetes: a machine-learning and regression analysis

Figure 1. Feature-importance rankings of TyG-related indices for TBS in patients with T2DM using Boruta, RF, and XGBoost. (A) Boruta ranking based on mean importance scores; (B) Random forest ranking based on permutation-derived %IncMSE; (C) XGBoost ranking based on gain; (D) Overall ranking based on the mean scaled importance across the three algorithms. %IncMSE: Percentage increase in mean squared error; RF: random forest; T2DM: type 2 diabetes mellitus; TBS: trabecular bone score; TyG: triglyceride-glucose index; XGBoost: extreme gradient boosting.

Metabolism and Target Organ Damage
ISSN 2769-6375 (Online)
Follow Us

Portico

All published articles are preserved here permanently:

https://www.portico.org/publishers/oae/

Portico

All published articles are preserved here permanently:

https://www.portico.org/publishers/oae/