Table 2

Performance of the ANNs in discriminating AD cases from normal controls. The analysis was carried out on all 4 neuropathologic variables registered in the original database of patients in ten separated experiments with different training and testing subsets. Linear Discriminant Analysis [LDA] results on the same subsets are shown for comparison.

Tr and Ts subsets

ANN

LDA


AD

Normal

Mean accuracy

AD

Normal

Mean accuracy


FF_Bp*(4 × 2)1a

100.00%

100.00%

100.00%

100.00%

87.50%

93.75%

FF_Bp(4 × 2)1b

100.00%

100.00%

100.00%

100.00%

91.67%

95.83%

FF_Bp(4 × 2)2a

100.00%

100.00%

100.00%

100.00%

72.73%

86.36%

FF_Bp(4 × 2)2b

100.00%

100.00%

100.00%

100.00%

88.89%

94.44%

FF_Bp(4 × 2)3a

100.00%

100.00%

100.00%

100.00%

87.50%

93.75%

FF_Bp(4 × 2)3b

100.00%

100.00%

100.00%

100.00%

83.33%

91.67%

FF_Bp(4 × 2)4a

100.00%

100.00%

100.00%

100.00%

72.73%

86.36%

FF_Bp(4 × 2)4b

100.00%

100.00%

100.00%

95.00%

100.00%

97.50%

FF_Bp(4 × 2)5a

100.00%

100.00%

100.00%

100.00%

91.67%

95.83%

FF_Bp(4 × 2)5b

100.00%

100.00%

100.00%

100.00%

75.00%

87.50%


Average

100.00%

100.00%

100.00%

99.50%

85.10%

92.30%


* Feed Forward Back Propagation Neural Network

Tr: Training set; TS: Testing set

Grossi et al. BMC Neurology 2007 7:15   doi:10.1186/1471-2377-7-15

Open Data