Skip to main content

Table 3 Comparison of the performance of wavelet bases on the MIAS dataset

From: Automatic detection of anomalies in screening mammograms

Wavelet basis§

Best feature combination

Sensitivity*(%)

Specificity†(%)

Overall‡classification rate (%)

Haar

S-h1

K-a2

K-a8

90.8

32.7

51.5

Db2

K-a3

  

93.9

14.1

39.9

Db4

K-h5

  

94.9

9.3

37.0

Db8

S-h3

S-a4

K-d4

91.8

27.3

48.2

Bior 1.5

K-h3

K-a1

K-a8

94.9

13.7

39.9

Bior 2.2

K-a2

  

94.9

14.1

40.3

Bior 2.8

S-d5

K-a4

 

94.9

27.3

49.2

Bior 3.7

S-d6

K-d8

K-a7

93.9

23.9

46.5

Bior 4.4

K-h6

K-a2

K-a5

93.9

16.1

41.3

Bior 5.5

K-a1

  

93.9

14.1

39.9

Bior 6.8

S-h3

K-h7

K-a3

94.9

22.0

45.5

  1. §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.
  2. *Sensitivity is defined as TP/(TP + FN).
  3. †Specificity is defined as TN/(TN + FP).
  4. ‡Classification rate is defined as (TP + TN)/(TP + TN + FP + FN).
  5. The best triplet feature was selected for each wavelet.