Our ultimate goal is to improve our ability to create and measure 3D models derived from cone-beam computed tomography (CBCT). Our main motivation is to improve quality and reduce costs in care of patients with craniomaxillofacial (CMF) deformities. The resulted innovations will also impact other fields. CMF deformities involve congenital and acquired deformities of the jaws and face. A large number of patients in the US and around the world suffered from CMF deformities. The evaluation of these patients includes an assessment of CMF form on 3D models that are traditionally generated from segmented spiral multi-slice CTs (MSCTs). These models are also used to plan their treatment. The purpose of segmentation is to separate different anatomical structures and to remove the artifacts on the CTs. Once 3D models are generated from the segmented CTs, anatomical and teeth landmarks are manually digitized for measurements. Finally, diagnosis and treatment planning are performed based on measurements. Although MSCT provides high- quality images and thus allows relatively fast and easy post processing, many concerns have been raised on excessive radiation exposure to patients. Therefore, more doctors are now using CBCT scanners in their offices. CBCT has less radiation and is inexpensive compared to the MSCT, but their use in generating 3D models is greatly limited by the poor image quality, i.e., low contrast / signal-to-noise ratio and artifacts. Thus, the existing automated segmentation algorithms developed for MSCT are incapable of practically segmenting CBCTs. The current solution to CBCT segmentation entails an arduous and lengthy process that involves labor-intensive manual editing of hundreds of slices. Besides, another arduous and inaccurate task in the assessment of CMF deformities is the digitization of anatomical landmarks on 3D models - the first step to quantify the deformities. Currently a typical 3D cephalometric and teeth analysis requires the manual digitization of more than 200 landmarks, which is time consuming and has limited accuracy. We hypothesize that the creation and measurement of high-quality 3D models can be significantly improved by developing innovative CBCT-friendly post processing tools. Therefore, in this renewal project, we propose to develop and validate a novel CBCT analysis platform to automate the process of CBCT segmentation and landmark digitization. The feasibility of our approaches has already been proven by our preliminary studies. Our innovative CBCT analysis platform will significantly improve the quality and reduce the cost of care to the individuals with CMF conditions. It will change our dental/CMF fields in effectively utilizing CBCT as a guide for on-the-fly diagnosis and treatment planning. With minimal user intervention, the computer will accurately and effectively do the work, which is currently artistically done by the labor-intensive human operators. The resulted innovations may also impact other fields in the future, e.g., orthopedic surgery and cardiovascular surgery where intraoperative whole-body CBCT is acquired for image-guided surgery and intervention.

Public Health Relevance

Cone-beam computed tomography (CBCT) is widely used in physician's offices for orthodontics, craniomaxillofacial (CMF) surgery, facial plastic surgery and dentistry, but its segmentation and landmark digitization have to be completed artistically by human operators, which is labor-intensive and with limited accuracy. We propose to develop and validate an innovative CBCT post processing system to automate the processes of CBCT segmentation and landmark digitization with minimal user intervention. The proposed system will significantly improve the quality and reduce the cost of care to the individuals with CMF conditions, and also change 1) the fields of orthodontics, CMF surgery and general dentistry in effectively utilizing CBCT as a guide for diagnosis and treatment planning, and 2) the fields of orthopedic surgery, general surgery, and cardiovascular surgery where the quality and the speed of intraoperative imaging is critical.

Agency
National Institute of Health (NIH)
Institute
National Institute of Dental & Craniofacial Research (NIDCR)
Type
Research Project (R01)
Project #
5R01DE022676-06
Application #
9417942
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Fischer, Dena
Project Start
2011-09-07
Project End
2021-01-31
Budget Start
2018-02-01
Budget End
2019-01-31
Support Year
6
Fiscal Year
2018
Total Cost
Indirect Cost
Name
Methodist Hospital Research Institute
Department
Type
DUNS #
185641052
City
Houston
State
TX
Country
United States
Zip Code
77030
Zhang, Yongqin; Shi, Feng; Cheng, Jian et al. (2018) Longitudinally Guided Super-Resolution of Neonatal Brain Magnetic Resonance Images. IEEE Trans Cybern :
Zhao, Miaoyun; Wang, Li; Chen, Jiawei et al. (2018) Craniomaxillofacial Bony Structures Segmentation from MRI with Deep-Supervision Adversarial Learning. Med Image Comput Comput Assist Interv 11073:720-727
Sutton, Peter H; Gateno, Jaime; English, Jeryl D et al. (2018) Both the Observer's Expertise and the Subject's Facial Symmetry Can Affect Anatomical Position of the Head. J Oral Maxillofac Surg :
Gateño, J; Jones, T L; Shen, S G F et al. (2018) Fluctuating asymmetry of the normal facial skeleton. Int J Oral Maxillofac Surg 47:534-540
Zhang, Xiaoyan; Kim, Daeseung; Shen, Shunyao et al. (2018) An eFTD-VP framework for efficiently generating patient-specific anatomically detailed facial soft tissue FE mesh for craniomaxillofacial surgery simulation. Biomech Model Mechanobiol 17:387-402
Yin, Q; Hung, S-C; Rathmell, W K et al. (2018) Integrative radiomics expression predicts molecular subtypes of primary clear cell renal cell carcinoma. Clin Radiol 73:782-791
Yin, Qingbo; Hung, Sheng-Che; Wang, Li et al. (2017) Associations between Tumor Vascularity, Vascular Endothelial Growth Factor Expression and PET/MRI Radiomic Signatures in Primary Clear-Cell-Renal-Cell-Carcinoma: Proof-of-Concept Study. Sci Rep 7:43356
Nie, Dong; Wang, Li; Trullo, Roger et al. (2017) Segmentation of Craniomaxillofacial Bony Structures from MRI with a 3D Deep-Learning Based Cascade Framework. Mach Learn Med Imaging 10541:266-273
Hughes, G N; Gateño, J; English, J D et al. (2017) There is variability in our perception of the standard head orientation. Int J Oral Maxillofac Surg 46:1512-1516
Wang, Jun; Wang, Qian; Peng, Jialin et al. (2017) Multi-task diagnosis for autism spectrum disorders using multi-modality features: A multi-center study. Hum Brain Mapp 38:3081-3097

Showing the most recent 10 out of 56 publications