Figure 3 | Scientific Reports

Figure 3

From: Unlocking the complete blood count as a risk stratification tool for breast cancer using machine learning: a large scale retrospective study

Figure 3

Explainability analysis. (A) Bar graph displaying Ridge Regression coefficients. (B) SHAP summary plot for the LightGBM model, showing the effect of each feature on BC risk prediction. Features are depicted in the order of importance. Pink dots are associated with women for which the corresponding feature shows a relatively higher value and blue dots with the opposite. Positions further to the right signify a greater predicted BC risk. NLR = neutrophils-to-lymphocytes ratio; RBC = red blood cells.

Back to article page