Fig 5. Classification of macrophage subtypes using SVM on morphological and motility parameters—Stratified 5-fold cross-validation plots for the SVM model.

(A) Schematics of the input and output of the machine learning model. SVM classification results showing the percentage distribution of the cells in a confusion matrix with columns being the predicted labels and rows being the actual labels. (B) Input features with morphological parameters (Eccentricity, Compactness, Cell Size). (C) Input features with motility parameters (Convex Hull Area of the trajectory, Speed, Persistence). (D) Input features with both morphological and motility parameters.