For each feature, a clear trend was observed: red dots (high feature values) clustered on the right (positive SHAP values) and blue dots (low feature values) on the left (negative SHAP values), demonstrating a positive association between the feature and the predicted risk of RA-ILD.
← all excerpts
A machine learning-based classification model for interstitial lung disease in rheumatoid arthritis.
2
—
—
The sentences
Variables demonstrating a trend toward significance on univariate analysis were subjected to LASSO regression, and feature variables were ultimately identified.