Deformable registration of lateral cephalogram and cone‐beam computed tomography image

Abstract

Purpose: This study aimed to design and evaluate a novel method for the registration of 2D lateral cephalograms and 3D craniofacial cone‐beam computed tomography (CBCT) images, providing patient‐specific 3D structures from a 2D lateral cephalogram without additional radiation exposure. Methods: We developed a cross‐modal deformable registration model based on a deep convolutional neural network. Our approach took advantage of a low‐dimensional deformation field encoding and an iterative feedback scheme to infer coarse‐to‐fine volumetric deformations. In particular, we constructed a statistical subspace of deformation fields and parameterized the nonlinear mapping function from an image pair, consisting of the target 2D lateral cephalogram and the reference volumetric CBCT, to a latent encoding of the deformation field. Instead of the one‐shot registration by the learned mapping function, a feedback scheme was introduced to progressively update the reference volumetric image and to infer coarse‐to‐fine deformations fields, accounting for the shape variations of anatomical structures. A total of 220 clinically obtained CBCTs were used to train and validate the proposed model, among which 120 CBCTs were used to generate a training dataset with 24k paired synthetic lateral cephalograms and CBCTs. The proposed approach was evaluated on the deformable 2D–3D registration of clinically obtained lateral cephalograms and CBCTs from growing and adult orthodontic patients. Results: Strong structural consistencies were observed between the deformed CBCT and the target lateral cephalogram in all criteria. The proposed method achieved state‐of‐the‐art performances with the mean contour deviation of 0.41±0.12 mm on the anterior cranial base, 0.48±0.17 mm on the mandible, and 0.35±0.08 mm on the maxilla, respectively. The mean surface mesh ranged from 0.78 to 0.97 mm on various craniofacial structures, and the LREs ranged from 0.83 to 1.24 mm on the growing datasets regarding 14 landmarks. The proposed iterative feedback scheme handled the structural details and improved the registration. The resultant deformed volumetric image was consistent with the target lateral cephalogram in both 2D projective planes and 3D volumetric space regarding the multicategory craniofacial structures. Conclusions: The results suggest that the deep learning‐based 2D–3D registration model enables the deformable alignment of 2D lateral cephalograms and CBCTs and estimates patient‐specific 3D craniofacial structures.

A review on AI-based medical image computing in head and neck surgery

“…This improvement expanded the learning scale from adjacent image layers to the whole image space, resulting also in the extraction of higher-level contextual features. Compared with the previous method, the accuracy and robustness were effectively improved, with the TRE of 0.92±0.67 mm and 1.02±0.67 mm in the original two datasets, respectively, and a test runtime of about 1.5 s. Zhang et al (2021a) proposed a CNN-based framework using iterative feedback (NRR IF ) for the deformable registration of 2D x-ray and 3D CBCT images. In this method, a low-dimensional encoding of the deformable field was performed to reduce the parametric space, and the registration was carried out from coarse to fine by iterative feedback.…”

Determination of internal target volume with Cine‐MRI sequence for prostate MRI‐Guided radiotherapy

“…To deal with the inter‐fractional deformation of CTV, we adapted the $$IT{V}_{ref}$$ to the shape of the CTV defined on positioning 3D‐MRI with deformable mesh registration (DMR). For the i ‐th fraction, DMR was first performed between $$CT{V}_{ref}$$ and CTV defined on the current positioning 3D‐MRI, 29 and the DVF establishing the inter‐fractional anatomy correspondence would be generated. Then, the DMR was performed between CTV ref and ITV ref , and the DVF that describes the motion pattern of the CTV was obtained.…”