Construction and validation of a risk prediction model for osteoporosis in COPD patients based on explainable machine learning.

In the stepwise multifactorial logistic regression (Table 3 ), variables such as platelet count, PT, creatinine level, hypertension, and glucocorticoid use did not reach statistical significance ( P > .05).

also in 111,027 other papers

Nevertheless, these variables retain potential contributory value to the model, indicating that although their independent effects may not be significant in the current model configuration, they may still exert substantial influence on the overall predictive performance of the model through synergistic interactions with other variables.

also in 2,774 other papers