TDSC-ABUS2023 Banner
Join

Tumor Detection, Segmentation and Classification Challenge on Automated 3D Breast Ultrasound (ABUS) 2023

Our challenge is now complete. However, you can still submit the signed document to tdscabus@gmail.com to access the training data. Please note that our validation and test data will be made available following the publication of our summary paper.
For further discussion, feel free to join us on Discord (https://discord.gg/KpHc3qxx) or QQ group 765737803.

[NEWS] 2023/08/30 - We are currently evaluating all submitted results; please check your email frequently for updates, as we will contact you immediately if there are any issues with your submission.

[NEWS] 2023/08/20 - TDSC-2023 Challenge test phase is now live! Follow submission instructions in the "Final Submit" page.
[NEWS] 2023/07/15 - TDSC-2023 Challenge validation phase is now live! Dataset has been emailed to participants. If not received, contact us. Follow submission instructions on our site.
[NEWS] 2023/04/07 - TDSC-ABUS 2023  is now open for registration! Remember to send the signed document to tdscabus@gmail.com for participation.

About

Breast cancer is one of the most common causes of death among women worldwide. Early detection helps in reducing the number of deaths. Automated 3D Breast Ultrasound is a newer approach for breast screening, which has many advantages over handheld mammography such as safety, speed, and higher detection rate of breast cancer. Thus it could prevail over the world in next several years. 
Tumor segmentation, classification and detection are three basic tasks in medical image analysis. These tasks are very challenging on 3D ABUS volumes for large tumor size and shape variation, irregular and ambiguous tumor boundaries, and low signal-to-noise ratio. Furthermore, there are rare open accessible ABUS datasets with well labeled tumor, which hinder the development of breast tumor segmentation, classification and detection systems.
Thus, we try to host the first Tumor Detection, Segmentation and Classification Challenge on Automated 3D Breast Ultrasound 2023 (Named TDSC-ABUS2023) to start a new research topic and make a solid benchmark for 3D ABUS image segmentation, classification and detection tasks. 
We have collected 200 3D volumes with refined both tumor boundaries and categories labeling from an experienced clinician, 100 for the training dataset, 70 for the closed testing dataset and 30 for the opened validated dataset. Dice, HD, are adopted as evaluation metrics for segmentation, accuracy and AUC are used as evaluation metrics for classification and  FROC is taken for detection. This challenge will also promote the breast cancer treatment, interactions between researchers and interdisciplinary communication. 

Task

Participants are required to Detect, Segment and Classify tumors in the 3D ABUS images.

Schedule

  • Registration Opens: March 28, 2023 (11:59 PM GMT)
  • Training Dataset Release: April 6, 2023 (11:59 PM GMT)
  • Validation Dataset Release & Open Validation Leaderboard Submission: July 15, 2023 (11:59 PM GMT)
  • Validation Leaderboard Submission Deadline: August 20, 2023 (11:59 PM GMT)
  • Opening of Docker and Short Paper Submission for Testing Phase: August 20, 2023 (11:59 PM GMT)
  • Docker and Short Paper Submission Deadline: August 30, 2023 (11:59 PM GMT)
  • Winner Announcement & Invitation to Speakers: October 8, 2023 (11:59 PM GMT)

Registration

Award

  1. Successful participation awards, which are electronic certificates, will be awarded to all teams that obtain valid test scores in the challenge leaderboard and complete technical paper submissions reviewed by the organizing committee.
  2. The top-1 team get best score on overall board will receive 300 dollars or electronic products with similar prices. The exquisite certificates will be awarded to all members of the Top-1 team.
  3. The top-1 team that win the first place on single task(Segmentation, Classification, Detection) board will receive 200 dollars or electronic products with similar prices. The exquisite certificates will be awarded to all members of the Top-1 team.

Citation

Any publication related to this challenge should reference our summary paper, which will be published immediately after the challenge concludes.