Automatic Segmentation of Lung Carcinoma Using 3D Texture Features in 18-FDG PET/CT

<table>  The average specificity for each algorithm. The majority of the tested algorithms had a higher specificity for the test data following the trend of the Dice coefficients with the exception of the MVLS method. </table>

International Journal of Molecular Imaging

fig8

Figure 8

Figure 8: Automatic Segmentation of Lung Carcinoma Using 3D Texture Features in 18-FDG PET/CT 