Journal
PHYSICS IN MEDICINE AND BIOLOGY
Volume 68, Issue 4, Pages -Publisher
IOP Publishing Ltd
DOI: 10.1088/1361-6560/acb4d7
Keywords
CBCT segmentation; direct segmentation; deformable image registration
Ask authors/readers for more resources
CBCT-based online adaptive radiotherapy requires accurate auto-segmentation, but DL-based direct segmentation of CBCT images is challenging due to poor quality and lack of well-labelled datasets. This study proposes a method using DIR and pseudo labels derived from deformed pCT contours for initial training, influencer volumes for defining the region of interest, and fine-tuning with a smaller set of true labels. Evaluation on nine patients shows that DL-based direct segmentation with influencer volumes improves performance to reach the level of DIR-based segmentation.
Cone-beam CT (CBCT)-based online adaptive radiotherapy calls for accurate auto-segmentation to reduce the time cost for physicians. However, deep learning (DL)-based direct segmentation of CBCT images is a challenging task, mainly due to the poor image quality and lack of well-labelled large training datasets. Deformable image registration (DIR) is often used to propagate the manual contours on the planning CT (pCT) of the same patient to CBCT. In this work, we undertake solving the problems mentioned above with the assistance of DIR. Our method consists of three main components. First, we use deformed pCT contours derived from multiple DIR methods between pCT and CBCT as pseudo labels for initial training of the DL-based direct segmentation model. Second, we use deformed pCT contours from another DIR algorithm as influencer volumes to define the region of interest for DL-based direct segmentation. Third, the initially trained DL model is further fine-tuned using a smaller set of true labels. Nine patients are used for model evaluation. We found that DL-based direct segmentation on CBCT without influencer volumes has much poorer performance compared to DIR-based segmentation. However, adding deformed pCT contours as influencer volumes in the direct segmentation network dramatically improves segmentation performance, reaching the accuracy level of DIR-based segmentation. The DL model with influencer volumes can be further improved through fine-tuning using a smaller set of true labels, achieving mean Dice similarity coefficient of 0.86, Hausdorff distance at the 95th percentile of 2.34 mm, and average surface distance of 0.56 mm. A DL-based direct CBCT segmentation model can be improved to outperform DIR-based segmentation models by using deformed pCT contours as pseudo labels and influencer volumes for initial training, and by using a smaller set of true labels for model fine tuning.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available