Table 1.
Textural indices.
Table 2.
Characteristics of patients.
Table 3.
Groups of highly correlated indices.
Table 4.
Textural indices correlation coefficient between themselves and with PET standard quantitative parameters (Pearson test).
Fig 1.
Variability between PET-EDGE and 40%SUVmax method.
Bland-Altman plot. (Solid blue line) Bias. (Dashed blue lines) Bias 95% confidence interval. (Dashed red lines) Difference 95% confidence interval.
Fig 2.
Variability between DAISNE and 40%SUVmax method.
Bland-Altman plot. (Solid blue line) Bias. (Dashed blue lines) Bias 95% confidence interval. (Dashed red lines) Difference 95% confidence interval.
Fig 3.
Variability between PET-EDGE and DAISNE method.
Bland-Altman plot. (Solid blue line) Bias. (Dashed blue lines) Bias 95% confidence interval. (Dashed red lines) Difference 95% confidence interval.
Fig 4.
Example of VOI delineating a tumor with the 3 segmentation methods.
(Turquoise blue) PET-EDGE segmentation method. (Sky blue) DAISNE segmentation method. (Dark blue) 40%SUVmax segmentation method. (Top left) FDG PET sagittal slice. (Top right) FDG PET transverse slice. (Bottom left) FDG PET frontal slice. (Bottom right) SUV histograms. With LIFEx software.
Table 5.
Inter-segmentation method variability (Friedman and Wilcoxon test p-values).
Table 6.
Inter-observer reproducibility (intra-class correlation coefficient).