Accuracy of 3D cephalometric measurements based on an automatic knowledge-based landmark detection algorithm

Gupta, Abhishek; Kharbanda, Om Prakash; Sardana, Viren; Balachandran, Rajiv; Sardana, Harish Kumar

doi:10.1007/s11548-015-1334-7

Cited by 54 publications

(47 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Normally, the number of landmarks used in the cephalometric analysis is directly based on the nature of a study, and there is no guarantee that more landmarks mean better outcomes. For example, Gupta et al used 20 landmarks in their study . By contrast, Li et al chose only 10 landmarks to conduct a landmarking work .…”

Section: Methodsmentioning

confidence: 99%

“…Therefore, some automatic landmarking methods have been proposed to relieve the workload of the surgeon and reduce the dependence of professional experience . These studies use different algorithms, such as registration‐based, Reeb‐graph‐based, active‐shape‐model‐based, and knowledge‐based method, which can reduce the analyzing time and improve the accuracy for automatic landmark localization in the 3D medical image. However, individual anatomical differences are remarkable, making such specifically developed algorithms lack of flexibility and be susceptible to data diversity.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic 3D landmarking model using patch‐based deep neural networks for CT image of oral and maxillofacial surgery

Kobayashi

Fan

et al. 2020

Robotics Computer Surgery

View full text Add to dashboard Cite

Background: Manual landmarking is a time consuming and highly professional work.Although some algorithm-based landmarking methods have been proposed, they lack flexibility and may be susceptible to data diversity. Methods: The CT images from 66 patients who underwent oral and maxillofacial surgery (OMS) were landmarked manually in MIMICS. Then the CT slices were exported as images for recreating the 3D volume. The coordinate data of landmarks were further processed in Matlab using a principal component analysis (PCA) method. A patch-based deep neural network model with a three-layer convolutional neural network (CNN) was trained to obtain landmarks from CT images. Results: The evaluating experiment showed that this CNN model could automatically finish landmarking in an average processing time of 37.871 seconds with an average accuracy of 5.785 mm. Conclusion: This study shows a promising potential to relieve the workload of the surgeon and reduces the dependence on human experience for OMS landmarking. K E Y W O R D S 3D cephalometry, automatic landmarking, convolutional neural network, machine learning, oral and maxillofacial surgery

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Automatic 3D landmarking model using patch‐based deep neural networks for CT image of oral and maxillofacial surgery

Kobayashi

Fan

et al. 2020

Robotics Computer Surgery

View full text Add to dashboard Cite

show abstract

“…With increasing clinical applications of CBCT imaging and growing implications of artificial intelligence (AI) such as automated 3D cephalometrics 24 and automated 3D airway analysis 16 in the craniofacial analysis, need for establishment of the correct reference plane for head orientation is required. Therefore, a study was conducted to evaluate the reliability and reproducibility of five methods of orientation of as-received CBCT images in the 3D space using landmark-based craniofacial reference planes.…”

Section: Introductionmentioning

confidence: 99%

Orientation of Cone-Beam Computed Tomography Image: Pursuit of Perfect Orientation Plane in Three Dimensions—A Retrospective Cross-Sectional Study

Balachandran

Kharbanda

Sennimalai

et al. 2019

Annals of the National Academy of Medical Sciences (India)

Self Cite

View full text Add to dashboard Cite

Objective This study aimed to evaluate the reproducibility of nine reference planes used in orientation of as-received cone-beam computed tomography (CBCT) images in all three dimensions. Materials and Methods The study was conducted on CBCT images of 15 adult subjects (mean age 21.2 ± 5.8 years). The anonymized CBCT images were oriented using five different methods created from nine reference planes by two experienced orthodontists. For each subject, pitch, yaw, and roll changes with five orientation methods were recorded twice by each observer. Statistical Analysis The inter- and intraobserver agreement was tested using intraclass correlation (ICC) and Bland–Altman plot. The intra- and interobserver error was analyzed using paired t-test. Analysis of variance and paired t-test were used to analyze the differences among the various pitch, roll, and yaw orientation planes. Results Inter- and intraobserver agreement (ICC, 0.9) was excellent for all the nine reference planes. The interobserver reliability showed statistically significant differences for four planes namely Frankfort horizontal plane constructed on right side (p = 0.014) and left side (p = 0.000), transorbital plane (p = 0.001), and midsagittal plane on top view (p = 0.036); however, the mean differences were clinically insignificant. Conclusion The landmark-based nine reference planes used in this study to orient CBCT images showed good reproducibility. Therefore, these reference planes can be used to orient CBCT images and can be incorporated into automated software.

show abstract

“…However, research purposes and surgery planning go beyond simple visualization, and therefore several 3D cephalometric tools have also been proposed to quantify linear and angular craniofacial measurements, transitioning from 2D to 3D analyses. 14 – 20 Most of these studies use CBCT to visualize a specific region, but still perform an overall overview of the patient using reformatted 2D images.…”

Section: Introductionmentioning

confidence: 99%

Challenges in measuring angles between craniofacial structures

et al. 2019

View full text Add to dashboard Cite

Objective: Three-dimensional (3D) angular measurements between craniofacial planes pose challenges to quantify maxillary and mandibular skeletal discrepancies in surgical treatment planning. This study aims to compare the reproducibility and reliability of two modules to measure angles between planes or lines in 3D virtual surface models. Methodology: Twenty oriented 3D virtual surface models de-identified and constructed from CBCT scans were randomly selected. Three observers placed landmarks and oriented planes to determine angular measurements of pitch, roll and yaw using (1) 3D pre-existing planes, (2) 3D planes created from landmarks and (3) lines created from landmarks. Inter- and intra-observer reproducibility and repeatability were examined using the Intra-Class Correlation (ICC) test. One observer repeated the measurements with an interval of 15 days. ANOVA was applied to compare the 3 methods. Results: The three methods tested provided statistically similar, reproducible and reliable angular measurements of the facial structures. A strong ICC varying from 0.92 to 1.00 was found for the intra-observer agreement. The inter-observer ICC varied from 0.84 to 1.00. Conclusion: Measurements of 3D angles between facial planes in a common coordinate system are reproducible and repeatable either using 3D pre-existing planes, created based on landmarks or angles between lines created from landmarks.

show abstract

Accuracy of 3D cephalometric measurements based on an automatic knowledge-based landmark detection algorithm

Abstract: Cephalometric measurements computed from automatic detection of landmarks on 3D CBCT image were as accurate as those computed from manual identification.

Cited by 54 publications

References 33 publications

Automatic 3D landmarking model using patch‐based deep neural networks for CT image of oral and maxillofacial surgery

Automatic 3D landmarking model using patch‐based deep neural networks for CT image of oral and maxillofacial surgery

Orientation of Cone-Beam Computed Tomography Image: Pursuit of Perfect Orientation Plane in Three Dimensions—A Retrospective Cross-Sectional Study

Challenges in measuring angles between craniofacial structures

Contact Info

Product

Resources

About