Using Convolutional Neural Network with Cheat Sheet and Data Augmentation to Detect Breast Cancer in Mammograms

Table 1

Summary of some methods used in breast cancer detection using CNN [13].

Author	Method	Database	Task	Metric/value(s)
Dhungel et al. [14]	Hybrid CNNa+level set	INbreast	Mass/classification	Accuracy (0.9) and sensitivity (0.98)
Dhungel et al. [15]	CRFc+CNN	INbreast and DDSMd	Lesion/segmentation	Dice score (0.89)
Singh et al. [16]	Conditional generative adversarial network and CNN	DDSM and Reus Hospital Spain dataset	Lesion/classification	Dice score (0.94) and Jaccard index (0.89)
Agarwal and Carson [17]	CNN (scratch based)	DDSM	Mass/calcifications	Accuracy (0.90)
Gao et al. [18]	Shallow-deep convolutional neural network CNN+ResNet	Mayo Clinic Arizona, INbreast	Lesion/classification	Accuracy (0.9) and AUC (0.92)
Hagos et al. [19]	Multi-input CNN	General Electric, Hologic, Siemens	Lesion/classification	AUC (0.93) and CPM (0.733)