Table 4: Comparison of ConvNet classifiers and visual features (color, shape, texture and topography) in terms of AUC.

From: Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent

MethodAUC
6-layer ConvNet0.9021 +/− 0.0097
3-layer ConvNet0.9018 +/− 0.0093
4-layer ConvNet0.8915 +/− 0.0093
Color and intensity features47,55,560.8711 +/− 0.0947
Color histograms47,560.8448 +/− 0.1047
Shape Index histogram570.8444 +/− 0.1065
Haralick features47,56,58.0.8385 +/− 0.0942
Topography and Graph-based features46,55,560.7998 +/− 0.1068