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

Table 12

Logic Based.

Reference	Descriptor	Image type	Number of images	Key findings
Beura et al. [95]	Two-dimensional discrete orthonormal -transform has been used for the feature extraction	Mammogram	—	Achieved Accuracy and AUC values on MIAS database are 98.3%, 0.9985.; Achieved Accuracy and AUC values on DDSM database are 98.8%, 0.9992.
Diz et al. [96]	GLCM	Mammogram	410	Their achieved Accuracy value is 76.60%
	GLRLM			Mean false positive value is 81.00%.
Zhang et al. [97]	133 features (mass based and content based)	Mammogram	400	Computer model has been created which is able to find a location that was not detected by trainee.
Ahmad and Yusoff [98]	Nine features selected	Biopsy	700	Achieved Sensitivity, Specificity, and Accuracy are 75.00%, 70.00%, and 72.00%, respectively.
Paul et al. [99]	Harlick texture feature	Histopathological	50	Their achieved Recall and Precision are 81.13% and 83.50%.
Chen et al. [100]	Dual-tree complex wavelet transform (DT-CWT) has been used for the feature extraction.	Mammogram	—	Achieved Received Operating Curve (ROC) 0.764.
Zhang et al. [101]	Curvelet Transform; GLCM CLBP	Histopathological	50	Random Subspace Ensemble (RSE) utilized.; Their achieved classification Accuracy is 95.22% where the previous Accuracy on this same database was 93.40%.