Skip to main content

Advertisement

Table 4 Results of nodule classification at wavelet second decomposition level

From: Automated system for lung nodules classification based on wavelet feature descriptor and support vector machine

Base Angle Sub-band Features Specificity Sensitivity Preciseness
Db1 HH2 Clpr-Ener 68.18% 68.18% 71.11%
Db2 LL Autc-Sent 36.36% 91.30% 64.44%
Db4 LL Autc-Ent 36.36% 95.45% 65.90%
Db1 45° LL Clsh-Ener 82.60% 68.18% 75.55%
Db2 45° LL Autc-Ener 73.91% 63.63% 68.88%
Db4 45° LL Autc-Sent 82.6% 59.09% 71.11%
Db1 90° LL Autc-Ener 86.95% 50% 68.88%
Db2 90° LL Autc-Ent 47.82% 95.45% 71.11%
Db4 90° LL Autc-Ent 54.16% 71.42% 62.22%
Db1 135° LL Autc-Ener 65.21% 63.63% 64.44%
Db2 135° LL Autc-Ent 60.86% 77.27% 68.88%
Db4 135° HH Autc-Sent 90.90% 56.21% 73.33%
  1. The bold data represent the best value obtained.