A fully automatic AI system for tooth and alveolar bone segmentation from cone-beam CT images

Cui, Zhiming; Fang, Yuheng; Mei, Lanzhuju; Zhang, Bojun; Yu, Bo; Liu, Jiameng; Jiang, Caiwen; Sun, Yuhang; Ma, Lei; Huang, Jiawei; Liu, Yang; Zhao, Yiqiang; Lian, Chunfeng; Ding, Zhongxiang; Zhu, Muhua; Shen, Dinggang

doi:10.1038/s41467-022-29637-2

Cited by 118 publications

(86 citation statements)

References 40 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mandibular dental implant planning requires detection or segmentation of the AB and MC. Cui et al proposed automatic tooth and alveolar bone segmentation on 3D CBCT images using the V-Net method which is a 3D fully CNN [12]. The accuracy of the alveolar bone segmentation in that study reached a Dice value of 94.5%.…”

Section: Introductionmentioning

confidence: 89%

“…The deep learning approach was initially implemented in dental radiology research [9]. Deep learning has been used to successfully detect bone radiography levels in panoramic radiographs [10], localize the MC on CBCT volume [6], classify teeth on CBCT images [11], segment AB on CBCT images [12], segment the mandibular cortical bone [13], MC [14] [15], tooth [12][16] [17], and inferior alveolar nerve [18] on CBCT images.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dental-YOLO: Alveolar Bone and Mandibular Canal Detection on Cone Beam Computed Tomography Images for Dental Implant Planning

et al. 2022

View full text Add to dashboard Cite

In planning a mandibular posterior dental implant, identifying the exact location of the alveolar bone (AB) and mandibular canal (MC) is essential to determine the height and width of the available bone. Cone beam computed tomography (CBCT) is a 3D imaging modality widely used for dental implant planning, which requires a lower radiation dose compared to medical CT and can provide cross-sectional image quality to visualize AB and MC. The radiologist carried out the AB and MC detection processes manually on each section of the CBCT image until the appropriate area was determined for bone measurement. This process is time consuming, and the measurement accuracy depends on the ability and experience of the radiologist. This study proposes an automatic and simultaneous detection system for AB and MC based on 2D grayscale CBCT images, that can simplify and expedite dental implant planning. We introduce Dental-YOLO, an efficient version of YOLOv4 specifically developed to detect AB and MC, with two-scale feature maps at low and high scales. The height and width of the available bone in the implant area were estimated by using the detected bounding box attributes. The AB and MC detection performances using Dental-YOLO reached a mean average precision of 99.46%. The two-way analysis of variance (ANOVA) test showed no difference in the bone height and width measurements produced by the proposed approach and manual measurement by radiologists. Our results suggest that the Dental-YOLO detection system could be helpful for dental implant surgery and presurgical treatment planning.INDEX TERMS Alveolar bone, CBCT, bone measurement, dental implant planning, mandibular canal, object detection, YOLO.

show abstract

Section: Introductionmentioning

confidence: 89%

Section: Introductionmentioning

confidence: 99%

Dental-YOLO: Alveolar Bone and Mandibular Canal Detection on Cone Beam Computed Tomography Images for Dental Implant Planning

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Moreover, a confidence-aware cascade segmentation module is designed in the second stage to segment each tooth. Recently, Cui et al (2022) proposed a multi-level morphology to guide the tooth segmentation model, which characterized the tooth shape from different angles of “point, line, and surface” and accurately extracted the patient’s dental crown and tooth root information.…”

Section: Related Workmentioning

confidence: 99%

Clinical tooth segmentation based on local enhancement

Zhang

Yang

et al. 2022

Front. Mol. Biosci.

View full text Add to dashboard Cite

The tooth arrangements of human beings are challenging to accurately observe when relying on dentists’ naked eyes, especially for dental caries in children, which is difficult to detect. Cone-beam computer tomography (CBCT) is used as an auxiliary method to measure patients’ teeth, including children. However, subjective and irreproducible manual measurements are required during this process, which wastes much time and energy for the dentists. Therefore, a fast and accurate tooth segmentation algorithm that can replace repeated calculations and annotations in manual segmentation has tremendous clinical significance. This study proposes a local contextual enhancement model for clinical dental CBCT images. The local enhancement model, which is more suitable for dental CBCT images, is proposed based on the analysis of the existing contextual models. Then, the local enhancement model is fused into an encoder–decoder framework for dental CBCT images. At last, extensive experiments are conducted to validate our method.

show abstract

“…Advances in digital dentistry are followed by increasing attempts to computerise certain routine clinical procedures, particularly the analysis of radiographs 23,24 . Artificial intelligence models for tooth and alveolar bone segmentation from cone-beam computed tomography images 23 , classification of cervical maturation degree and pubertal growth spurts from lateral cephalometric radiographs 24 would reduce the need for manual evaluation of radiographic images and contribute to treatment efficiency. However, to accurately visualise the tooth structure, a higher resolution 3D imaging technique is needed than has been used so far.…”

Section: Introductionmentioning

confidence: 99%

Microcracks as tooth structural element: revealed by X-ray tomography and machine learning

Dumbryte

Narbutis

Vailionis

et al. 2022

Preprint

View full text Add to dashboard Cite

Although teeth microcracks have long been considered more of an aesthetic problem, their exact role in tooth structure and impact on its functionality is still unknown. The possibility to use X-ray micro-computed tomography for three-dimensional (3D) non-destructive tooth visualization has shown unique morphological and structural features of microcracks that previously could not be detected with two-dimensional analysis techniques. We present 3D evaluation of four healthy (2 with and 2 without visible microcracks on the buccal surface) extracted human teeth, scanned with ZEISS Xradia 520 Versa and segmented with convolutional neural network to identify enamel, dentin, and cracks. An intricate star-shaped network of microcracks was found to cover most of the inner tooth with the main crack planes arranged radially in two almost perpendicular directions, disclosing the crack as a planar rather than a threaded structure. The interconnected network of microcracks detected inside a healthy tooth suggests that the cracks could be considered as one of the structural and possibly functional (i.e. redistribution of forces) elements of the tooth, relevant for its integrity and functionality.

show abstract

A fully automatic AI system for tooth and alveolar bone segmentation from cone-beam CT images

Cited by 118 publications

References 40 publications

Dental-YOLO: Alveolar Bone and Mandibular Canal Detection on Cone Beam Computed Tomography Images for Dental Implant Planning

Dental-YOLO: Alveolar Bone and Mandibular Canal Detection on Cone Beam Computed Tomography Images for Dental Implant Planning

Clinical tooth segmentation based on local enhancement

Microcracks as tooth structural element: revealed by X-ray tomography and machine learning

Contact Info

Product

Resources

About