Abstract
The lengthy time needed for manual landmarking has delayed the widespread adoption of three-dimensional (3D) cephalometry. We here propose an automatic 3D cephalometric annotation system based on multi-stage deep reinforcement learning (DRL) and volume-rendered imaging. This system considers geometrical characteristics of landmarks and simulates the sequential decision process underlying human professional landmarking patterns. It consists mainly of constructing an appropriate two-dimensional cutaway or 3D model view, then implementing single-stage DRL with gradient-based boundary estimation or multi-stage DRL to dictate the 3D coordinates of target landmarks. This system clearly shows sufficient detection accuracy and stability for direct clinical applications, with a low level of detection error and low inter-individual variation (1.96 ± 0.78 mm). Our system, moreover, requires no additional steps of segmentation and 3D mesh-object construction for landmark detection. We believe these system features will enable fast-track cephalometric analysis and planning and expect it to achieve greater accuracy as larger CT datasets become available for training and testing.
Highlights
The lengthy time needed for manual landmarking has delayed the widespread adoption of threedimensional (3D) cephalometry
Landmark localization accuracy. 3D coordinate values of landmarks determined by human experts and the experimental values obtained by our proposed method were independently produced and compared in terms of 3D mean distance between them
To determine possible differences in 3D landmark prediction based on the number of deep reinforcement learning (DRL) passes, we tried four passes of DRL inferencing for each test group landmark
Summary
The lengthy time needed for manual landmarking has delayed the widespread adoption of threedimensional (3D) cephalometry. We here propose an automatic 3D cephalometric annotation system based on multi-stage deep reinforcement learning (DRL) and volume-rendered imaging This system considers geometrical characteristics of landmarks and simulates the sequential decision process underlying human professional landmarking patterns. It consists mainly of constructing an appropriate two-dimensional cutaway or 3D model view, implementing single-stage DRL with gradient-based boundary estimation or multi-stage DRL to dictate the 3D coordinates of target landmarks. Deep learning methods using convolutional neural networks, predict a spatial location by a single-shot decision based on training results from huge amounts of labelled data This decision-making process cannot be properly adapted to complex structures with variation/deformation.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
