Prospective validation of 2D CNN auto-segmentation for cervical cancer radiotherapy planning

Auto-segmentation for cervical cancer CT achieved mean Dice scores of 0.92 (femur), 0.89 (bowel bag), 0.88 (bladder) and reduced contour time from 77 min to 5 sec (p<0.001) in a prospective study.

• Among target volumes, inguinal nodal CTV achieved highest agreement (DSC 0.77), while pelvic nodal CTV was lower (DSC 0.60).
• Clinical acceptability was rated Likert III–IV for key pelvic organs; performance was lower for complex structures like duodenum and pelvic vessels.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.