Analysis of the Cluster Prominence Feature for Detecting Calcifications in Mammograms

Table 3

Comparison between our methodology and state-of-the-art methods.

Compared methods	Dataset	Set-up	Accuracy	Sensitivity	FPi
Proposed strategy	Hosp. Irapuato	A	0.9189	0.9090	5.4054
PSOWNN [15]	Clinical	A	0.9367	0.9416	1.9006
DEOWNN [25]	MIAS	A	0.9353	0.9690	5.9754
Hough transform [18]	BCDR	A	0.9326	0.9178	3.9999
CALMA-ANN [19]	CALMA	A	0.9200	0.9200	4.9627
CAD-PPJ [20]	U of M.	A	0.8914	0.8500	1.7100
Texture feature + SLDA [26]	DDSM	S	0.8700	0.9333	10.000
Level set [27]	U. of M.	S	0.8500 ± 0.0200	—	—
Active contour [27]	U. of M.	S	0.8700 ± 0.0200	—	—