Skip to main content

Table 4 The confusion matrices use the area feature, PCA and KPCA approaches to classify different gaits

From: Quantification and recognition of parkinsonian gait from monocular video imaging using kernel-based principal component analysis

  Area feature PCA KPCA
  Non-PD Drug-Off Drug-On Non-PD Drug-Off Drug-On Non-PD Drug-Off Drug-On
Non-PD 9 0 3 10 1 1 10 0 2
Drug-Off 1 8 3 2 7 3 2 10 0
Drug-On 6 5 1 3 1 8 2 1 9
Predicted/actual 9/12 8/12 1/12 10/12 7/12 8/12 10/12 10/12 9/12
Accuracy 75% 66.67% 8.33% 83.33% 58.33% 66.67% 83.33% 83.33% 75%
Average accuracy 50.16%    69.44%    80.51%