Automatic detection of anomalies in screening mammograms

BMC Medical Imaging

Table 4 Performance of sequential classifiers using the DDSM database

Wavelet basis^§	Classified normal		Classified suspicious		Confidence level (%)
	Actually normal	Actually suspicious	Actually normal	Actually suspicious	C _even *(N)*	C _even *(S)*
Haar	390	5	675	644	97.9	61
Bior 3.7	268	2	407	642	98.8	72.1
Bior 2.2	103	4	304	638	94	77.5
Bior 6.8	131	7	173	631	91.9	85.7
Db 2	87	10	86	621	84.1	92.2
Bior 5.5	20	6	67	615	67	93.8
Bior 1.5	21	8	45	607	61.5	95.7
Db 8	15	11	30	596	45.5	97

§Wavelet basis notation: Dbn where n describes the number of coefficients used in the wavelet. Db2 encodes polynomials with two coefficients, i.e. constant and linear components. Biorm.n where n describes the order for decompositiona and m is the order used for reconstruction.
Suspicious images were passed to the next classifier in the chain.

ISSN: 1471-2342