Skip to main content

Table 1 Performances of KNN with Euclidean, Cosine, and Minkowski distance metrics

From: A study of hepatic fibrosis staging methods using diffraction enhanced imaging

Distance metric Accuracy Sensitivity Specificity Training time(s) Prediction speed (obs/s)
Euclidean 99.21% 99.38% 98.54% 2.4916 18,000
Cosine 99.13% 99.22% 98.80% 2.503 18,000
Minkowski 99.23% 99.39% 98.54% 58.548 570