Dehazing Echocardiography Challenge 2025 Banner
Join

Dehazing Echocardiography Challenge 2025

Echocardiography is one of the most significant advances in cardiac imaging due to its real-time capabilities and cost-effectiveness. It enables cardiologists to assess the basic functioning of the heart and detect abnormalities that serve as critical markers for cardiovascular diseases. However, transthoracic echocardiograms are often affected by various noise sources that degrade image quality and limit interpretability. This highlights the urgent need for enhancing echocardiographic image quality. Dehazing echocardiography for difficult-to-image patients using advanced computational algorithms has the potential to significantly enhance the accuracy of medical diagnoses and improve the performance of AI-assisted algorithms.

We propose to hold the challenge of dehazing echocardiography in conjunction with MICCAI 2025. The challenge will focus on dehazing and enhancing echocardiography for patients who are typically more difficult to image. To support this, we will provide a dataset consisting of 4500 four-chamber echocardiography images from 75 easy-to-image patients and 2400 four-chamber echocardiography images from 40 difficult-to-image patients in the challenge.

References:

  1. T.S.W. Stevens, F.C. Meral, J. Yu, I.Z. Apostolakis, J.-L. Robert, R.J.G. Van Sloun, Dehazing Ultrasound Using Diffusion Models, IEEE Trans. Med. Imaging 43 (2024) 3546–3558.
  2. L. Huang, J. Zhou, J. Jiao, S. Zhou, C. Chang, Y. Wang, Y. Guo, Standardization of ultrasound images across various centers: M2O-DiffGAN bridging the gaps among unpaired multi-domain ultrasound images, Medical Image Analysis 95 (2024) 103187.
  3. J. Jiao, J. Zhou, X. Li, M. Xia, Y. Huang, L. Huang, N. Wang, X. Zhang, S. Zhou, Y. Wang, Y. Guo, USFM: A universal ultrasound foundation model generalized to tasks and organs towards label efficient image analysis, Medical Image Analysis 96 (2024) 103202.
  4. M. Xia, H. Yang, Y. Qu, Y. Guo, G. Zhou, F. Zhang, Y. Wang, Multilevel structure-preserved GAN for domain adaptation in intravascular ultrasound analysis, Medical Image Analysis 82 (2022) 102614.
  5. L. Huang, Z. Zhou, Y. Guo, Y. Wang, A stability-enhanced CycleGAN for effective domain transformation of unpaired ultrasound images, Biomedical Signal Processing and Control 77 (2022) 103831.
  6. Z. Zhou, Y. Guo, Y. Wang, Ultrasound deep beamforming using a multiconstrained hybrid generative adversarial network, Medical Image Analysis 71 (2021) 102086.