Involvement of Machine Learning for Breast Cancer Image Classification: A Survey

Table 11

Convolutional Neural Network.

Reference	Descriptor	Image type	Number of images	Key findings
Jiang et al. [89]	Global Features	Mammogram	—	Image preprocessing was performed to enhance tissue characteristics.
				Transfer learning was performed and obtained AUC was 0.88 whereas when the system learned from scratch, the best ROC is 0.82.
Suzuki et al. [90]	Global Features	Mammogram	198	The achieved sensitivity 89.90%.
				Transfer learning techniques have been utilized.
Qiu et al. [91]	Global Features	Mammogram	270	Average achieved Accuracy is 71.40%.
Samala et al. [92]	Global Features	—	92	They utilized Deep Learning CNN (DLCNN) and CNN models for classification.
				The AUC of CNN and DLCNN model is 0.89 and 0.93, respectively.
Sharma and Preet [84]	Global Features	Mammogram	607	Transfer learning and ensemble techniques utilized.
				When using ensemble techniques the soft voting method has been used.
				The best ROC score is 0.86.
Kooi et al. [93]	Global and Local features	Mammogram	44090	Transfer learning method utilized (VGG model).
Geras et al. [94]	Global Features	Mammogram	102800	They investigated the relation of the Accuracy with the database size and image size.
Arevalo et al. [82]	Global Features	Mammogram	736	The best ROC value was 0.822.