Fully Automated Delineation of Gross Tumor Volume for Head and Neck Cancer on PET-CT Using Deep Learning: A Dual-Center Study

Figure 4

Another typical example of HNC tumor segmentation with low accuracy. The dice similarity coefficient (DSC) was 0.408. (a) PET image coregistered to CT. (b) CT image. (c) Automatic segmentation result presented on the fused PET-CT image (green line). (d) Gold standard of gross tumor volume, drawn on the fused PET-CT image (red line).

(a)

(b)

(c)

(d)