DeepOPG: Improving Orthopantomogram Finding Summarization with Weak Supervision

@article{Hsu2021DeepOPGIO,
  title={DeepOPG: Improving Orthopantomogram Finding Summarization with Weak Supervision},
  author={Tzu-Ming Harry Hsu and Yingni Wang},
  journal={ArXiv},
  year={2021},
  volume={abs/2103.08290}
}
Clinical finding summaries from an orthopantomogram, or a dental panoramic radiograph, have significant potential to improve patient communication and speed up clinical judgments. While orthopantomogram is a first-line tool for dental examinations, no existing work has explored the summarization of findings from it. A finding summary has to find teeth in the imaging study and label the teeth with several types of past treatments. To tackle the problem, we develop DeepOPG that breaks the… 

OdontoAI: A human-in-the-loop labeled data set and an online platform to boost research on dental panoramic radiographs

This study addresses the construction of a public data set of dental panoramic radiographs, using predictions from deep neural networks as provisional labels, later verified by human annotators and demonstrated a 51% labeling time reduction using HITL, saving more than 390 continuous working hours.

BDU‐net: Toward accurate segmentation of dental image using border guidance and feature map distortion

A dual subnetworks structure based on border guidance and feature map distortion, called BDU‐net, is proposed, which embeds the Disout method with the capability to distort features into the encoder to obtain multi‐scale feature maps with excellent generalization.

References

SHOWING 1-10 OF 21 REFERENCES

Deep Instance Segmentation of Teeth in Panoramic X-Ray Images

A segmentation system based on mask region-based convolutional neural network to accomplish an instance segmentation of the teeth is proposed, which is the first system that detects and segment each tooth in panoramic X-ray images.

Tooth detection and numbering in panoramic radiographs using convolutional neural networks.

The performance of the proposed computer-aided diagnosis solution is comparable to the level of experts and has the potential for practical application and further evaluation for automated dental radiograph analysis.

ToothNet: Automatic Tooth Instance Segmentation and Identification From Cone Beam CT Images

This method is the first to use neural networks to achieve automatic tooth segmentation and identification from CBCT images and encodes teeth spatial relationships as an additional feature input in the identification task, which helps to remarkably improve the identification accuracy.

Accurate Segmentation of Dental Panoramic Radiographs with U-NETS

An FCN based on the U-Net architecture is applied for the challenging task of semantic segmentation of dental panoramic radiographs and general tricks for improving segmentation performance are discussed.

Classification of teeth in cone-beam CT using deep convolutional neural network

Panoramic radiography can be an effective diagnostic tool adjunctive to oral examinations in the national health checkup program

Panoramic radiography can provide information that is difficult to obtain in oral examinations, such as root caries, furcation involvement, and subgingival calculus, which are factors that can directly affect the survival rate of teeth.

Deep Residual Learning for Image Recognition

This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.

Single‐image Tomography: 3D Volumes from 2D Cranial X‐Rays

This work shows that recent deep learning‐based convolutional neural networks can solve the task of inverting 2D x‐ray images into 3D volumes and suggests firstly to learn a coarse, fixed‐resolution volume which is then fused in a second step with the input x‐rays into a high‐ resolution volume.

Interpretation of panoramic radiographs.

How osseous structures, soft tissues, air spaces and ghost shadows contribute to the final panoramic image is described, which is recommended to ensure that critical findings are not overlooked.