DENTEX CHALLENGE 2023

Dental Enumeration and Diagnosis on Panoramic X-rays


FINAL DENTEX Workshop: Schedule & Joining Details

We have come to the conclusion of the Dental Enumeration and Diagnosis on Panoramic X-rays Challenge (DENTEX). We are now ready for the final workshop.

Join us on Sunday, October 8th, from 2 p.m. (Vancouver time) at MICCAI. This will be a hybrid event, accommodating both in-person (Oceanview Suite 1) and online attendees. Below is the detailed schedule, along with the Zoom link for our virtual participants.

  • 1400: Introduction to DENTEX by the organizers
  • 1410: Presentation by the team Radboud_ISMI
  • 1420: Presentation by the team teethseg
  • 1430: Presentation by the team Chohotech
  • 1440: Presentation by the team sjtu-seiee-426
  • 1450: Open Forum for Discussion and Feedback on DENTEX

Zoom link to join: https://uzh.zoom.us/j/62932371733?pwd=a1hDSXB1eURya0I2VjYrZFJRN3hkQT09

We're grateful for your invaluable contributions to DENTEX and look forward to the workshop.


FINAL ANNOUNCEMENTS

The DENTEX challenge is over. Thank you so much for your participation. The results are presented at the DELTA Workshop during MICCAI 2023 on October 8 in Vancouver, Canada.

Following a thorough evaluation, we have determined the final rankings. We are excited to announce the teams that will be recognized and awarded at MICCAI 2023:

  1. sjtu-seiee-426
  2. Chohotech
  3. sdent

Honorable Mentions: Ai-Align & huikai

The challenge results will be summarized and published in a journal manuscript. All participants are invited to co-author our challenge publication! We also invite all participants to prepare a 5-minute presentation or video to showcase and discuss their methods during the MICCAI workshop. 

To ensure your inclusion in our paper and/or to present your results at the MICCAI workshop, please:

  1. Complete the contact form.
  2. Draft a short paper (a minimum of four paragraphs) detailing your algorithm. You can either upload it to ArXiv or email it to us.

Challenge Paper Submission Guidance:

  1. Method Description: A brief 3–4 line explanation, though elaboration is welcomed.
  2. *Data: *Discuss the data used, preprocessing, augmentation, ROI treatment, image sampling, image orientation, voxel size treatment, and patches. Additionally, address any variations in results due to altered or modified data.
  3. Training Strategies: Detail the amount and type of resources utilized. For instance, specify the type and number of GPUs, whether an internal cluster or a cloud provider was used, and if a high-performance computing cluster aided in training.
  4. Acknowledgments and Availability: Share the status of Docker, code, and data availability and link any references.

Please Note: Meeting these criteria is essential for co-authorship. All participants, upon submission of the short paper, are eligible for co-authorship on the challenge paper, regardless of whether they've been awarded.

  • A single submission allows for a maximum of three co-authorships for the challenge paper. Should top submissions have more contributors, additional authors can be added upon request, provided there's a valid justification aligned with the ICMJE authorship guidelines.
  • Teams are welcome to individually submit their results without any restrictions from publication embargoes.

BACKGROUND

Panoramic X-rays are widely used in dental practice to provide a comprehensive view of the oral cavity and aid in treatment planning for various dental conditions. However, interpreting these images can be a time-consuming process that can distract clinicians from essential clinical activities. Moreover, misdiagnosis is a significant concern, as general practitioners may lack specialized training in radiology, and communication errors can occur due to work exhaustion.

In recent years, advancements in artificial intelligence (AI) have paved the way for automated dental radiology analysis. However, developing automated algorithms for panoramic X-ray analysis is challenging due to variations in anatomy and the lack of publicly available annotated data. Despite these challenges, the potential benefits of utilizing AI in dental radiology analysis cannot be ignored, as it can significantly improve treatment outcomes and patient satisfaction. Therefore, there is a growing need for research to explore and develop effective AI algorithms for dental radiology .

DENTEX

To address this limitation, we present the Dental Enumeration and Diagnosis on Panoramic X-rays Challenge (DENTEX), organized in conjunction with the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) in 2023. The primary objective of this challenge is to develop algorithms that can accurately detect abnormal teeth with dental enumeration and associated diagnosis. This not only aids in accurate treatment planning but also helps practitioners carry out procedures with a low margin of error.
The challenge provides three types of hierarchically annotated data and additional unlabeled X-rays for optional pre-training. The annotation of the data is structured using the Fédération Dentaire Internationale (FDI) system. The first set of data is partially labeled because it only includes quadrant information. The second set of data is also partially labeled but contains additional enumeration information along with the quadrant. The third data is fully labeled because it includes all quadrant-enumeration-diagnosis information for each abnormal tooth, and all participant algorithms will be benchmarked on the third data. 
DENTEX aims to provide insights into the effectiveness of AI in dental radiology analysis and its potential to improve dental practice by comparing frameworks that simultaneously point out abnormal teeth with dental enumeration and associated diagnosis on panoramic dental X-rays.

Fig. 1. A desired output from a final algorithm, illustrating well-defined bounding boxes for each abnormal tooth. *The corresponding quadrant (Q), enumeration (N), and diagnosis (D) labels are also displayed.*