Automatic reconstruction method for high-contrast panoramic image from dental cone-beam CT data

Yun, Zhaoqiang; Yang, Shuo; Huang, Erliang; Zhao, Lei; Yang, Wei; Feng, Qianjin

doi:10.1016/j.cmpb.2019.04.024

Cited by 35 publications

(35 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To relieve these data imbalance problems, reduce the computational load, and decrease the learning complexity, it is necessary to delimit a small and valid region from a dental CBCT image for segmentation. Yun et al [23] proposed the detection of the dental arch in dental CBCT images by thresholding operations and projection analyses on the maximum intensity projection (MIP) image. Similarly, we delimited the valid region for segmentation through a MIP-based method.…”

Section: B Preprocessingmentioning

confidence: 99%

Automatic Segmentation of Individual Tooth in Dental CBCT Images From Tooth Surface Map by a Multi-Task FCN

Chen

Yun

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Accurate and automatic segmentation of individual tooth is critical for computer-aided analysis towards clinical decision support and treatment planning. Three-dimensional reconstruction of individual tooth after the segmentation also plays an important role in simulation in digital orthodontics. However, it is difficult to automatically segment individual tooth in cone beam computed tomography (CBCT) images due to the blurring boundaries of neighboring teeth and the similar intensities between teeth and mandible bone. In this work, we propose the use of a multi-task 3D fully convolutional network (FCN) and marker-controlled watershed transform (MWT) to segment individual tooth. The multi-task FCN learns to simultaneously predict the probability of tooth region and the probability of tooth surface. Through the combination of the tooth probability gradient map and the surface probability map as the input image, MWT is used to automatically separate and segment individual tooth. Twenty-five dental CBCT scans are used in the study. The average Dice similarity coefficient, Jaccard index, and relative volume difference are 0.936 (±0.012), 0.881 (±0.019), and 0.072 (±0.027), respectively, and the average symmetric surface distance is 0.363 (±0.145) mm for our method. The experimental results demonstrate that the multi-task 3D FCN combined with MWT can segment individual tooth of various types in dental CBCT images.INDEX TERMS Individual tooth segmentation, dental CBCT, deep learning, marker-controlled watershed transform.

show abstract

Section: B Preprocessingmentioning

confidence: 99%

Automatic Segmentation of Individual Tooth in Dental CBCT Images From Tooth Surface Map by a Multi-Task FCN

Chen

Yun

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…Because dentomaxillofacial anatomic structures have different xray absorption characteristics, each structure has its own characteristic range of pixel values. 19 This principle has been utilized to design filters that can enhance visibility of specific structures. These state-of-the-art MDs with dental configurations apply significant amounts of image processing in real time to ensure that image quality and visualization are optimal at any moment.…”

Section: Statement Of Clinical Relevancementioning

confidence: 99%

The influence of image display systems on observers’ preference for visualizing subtle dental radiographic abnormalities

Shujaat

Letelier

Grauwe

et al. 2021

Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology

View full text Add to dashboard Cite

Objective. The objectives of this study were to assess observers' preference for standard screens (SSs) or medical displays (MDs) in visualizing difficult-to-diagnose radiographic dental abnormalities and their preference for dental filter tools when utilized with MD systems. Study Design. A retrospective data set of 60 in vivo radiographs consisting of intraoral (n = 20), panoramic (n = 20), and cone beam computed tomography (n = 20) images was created. Three image display monitors, including an SS, an MD, and an MD with 3 dental filter configurations (boneÀlow density enhancement filter, toothÀhigh density enhancement filter, and a combined filter representing regular MD), were utilized to assess 4 observers' monitor preferences in detecting radiographically subtle dental abnormalities. The data were analyzed by using binomial distribution. A P value .05 was considered statistically significant. Results. Although observers expressed preference for MD for visualizing some abnormalities when examining intraoral and panoramic radiographs, MD was not preferred for detection of any abnormalities with cone beam computed tomography. There were no significant differences in preference for SS or MD overall (P .2024). Observers expressed significant preference for the filters in visualizing all but 2 abnormalities (P .0252). Conclusions. The use of MD monitors enabled with dental filter tools may be preferred for visualizing certain subtle abnormalities.

show abstract

“…Such synthesis is valid since CBCT scans contain full 3D information of cavity, while panoramic radiographs are the 2D projections of them. Several previous work has demonstrated promising results for high quality synthesis, and in our work, we employ the method of Yun et al [16] for building our dataset. Our in-house dataset contains 23 pairs of 3D CBCT scans and panoramic radiographs, each with a resolution ranging form 0.250 mm to 0.434 mm.…”

Section: Datasetmentioning

confidence: 99%

X2Teeth: 3D Teeth Reconstruction from a Single Panoramic Radiograph

Liu

Song

Yang

et al. 2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

3D teeth reconstruction from X-ray is important for dental diagnosis and many clinical operations. However, no existing work has explored the reconstruction of teeth for a whole cavity from a single panoramic radiograph. Different from single object reconstruction from photos, this task has the unique challenge of constructing multiple objects at high resolutions. To conquer this task, we develop a novel ConvNet X2Teeth that decomposes the task into teeth localization and single-shape estimation. We also introduce a patch-based training strategy, such that X2Teeth can be end-to-end trained for optimal performance. Extensive experiments show that our method can successfully estimate the 3D structure of the cavity and reflect the details for each tooth. Moreover, X2Teeth achieves a reconstruction IoU of 0.681, which significantly outperforms the encoder-decoder method by 1.71× and the retrieval-based method by 1.52×. Our method can also be promising for other multi-anatomy 3D reconstruction tasks.

show abstract

Automatic reconstruction method for high-contrast panoramic image from dental cone-beam CT data

Cited by 35 publications

References 24 publications

Automatic Segmentation of Individual Tooth in Dental CBCT Images From Tooth Surface Map by a Multi-Task FCN

Automatic Segmentation of Individual Tooth in Dental CBCT Images From Tooth Surface Map by a Multi-Task FCN

The influence of image display systems on observers’ preference for visualizing subtle dental radiographic abnormalities

X2Teeth: 3D Teeth Reconstruction from a Single Panoramic Radiograph

Contact Info

Product

Resources

About