Program

Program

Instructions for presenters - click here

08:00 – 08:05 

Welcome and opening remarks 

08:05 – 08:35 

Oral Session 1 (2 talks - 15min each) 

  • 003: Subject-Specific Lesion Generation and Pseudo-Healthy Synthesis for Multiple Sclerosis Brain Images, Basaran, Qiao, Matthews, Bai (In-person) (code) 

  • 008: HealthyGAN: Learning from Unannotated Medical Images to Detect Anomalies Associated with Human Disease, Rahman Siddiquee, Shah, Wu, Chong, Schwedt, Li (code) 

08:35 – 09:05 

Poster Highlights (15 talks - 2min each) 

  • 003: Subject-Specific Lesion Generation and Pseudo-Healthy Synthesis for Multiple Sclerosis Brain Images, Basaran, Qiao, Matthews, Bai (In-person) (code) 

  • 004: Generating Artificial Artifacts for Motion Artifact Detection in Chest CT, van der Ham, Latisenko, Tsiaousis, van Tulder (code) 

  • 005: Probabilistic Image Diversification to Improve Segmentation in 3D Microscopy Image Data, Eschweiler, Schock, Stegmaier (code) 

  • 006: Pathology synthesis of 3D consistent cardiac MR images using 2D VAEs and GANs, Amirrajab, Lorenz, Weese, Pluim, Breeuwer (code) 

  • 008: HealthyGAN: Learning from Unannotated Medical Images to Detect Anomalies Associated with Human Disease, Rahman Siddiquee, Shah, Wu, Chong, Schwedt, Li (code) 

  • 010: Bi-directional Synthesis of Pre- and Post-contrast MRI via Guided Feature Disentanglement, Xue, Dewey, Zuo, Han, Carass, Duan, Remedios, Pham, Saidha, Calabresi, Prince (video) 

  • 011: Morphology-preserving Autoregressive 3D Generative Modelling of the Brain, Tudosiu, Graham, Fernandez, Yang, Appleyard, Novati, Mehra, Vella, Cardoso, Nachev, Ourselin, Borges, Pinaya (code) 

  • 012: Can segmentation models be trained with fully synthetically generated data? Fernandez, Pinaya, Borges, Tudosiu, Graham , Vercauteren , Cardoso 

  • 013: Multimodal super resolution with dual domain loss and gradient guidance, Krishnan, UPENDRA, Pramanik, Song, Carano 

  • 014: Brain Lesion Synthesis via Progressive Adversarial Variational Auto-Encoder, Huo, Vakharia, Wu, Sharan, Ko, Ourselin, Sparks 

  • 016: Contrastive Learning for Generating Optical Coherence Tomography Images of the Retina, Kaplan, Lensu (code) 

  • 017: A novel method combining global and local assessments to evaluate CBCT-based synthetic CTs, Sargeant, Green, Shortall, Chuter, Xu, Thill, O'Connell, McWilliam 

  • 018: SuperFormer: Volumetric Transformer Architectures for MRI Super-Resolution, Forigua, Escobar, Arbeláez (code) 

  • 019: Evaluating the Performance of StyleGAN2-ADA on Medical Images, Woodland, Wood, Anderson, Kundu, Lin, Koay, Odisio, Chung, Kang, Venkatesan, Yedururi, De, Lin, Patel, Brock 

  • 021: Backdoor Attack is A Devil in Federated GAN-based Medical Image Synthesis, Jin, Li (code) 

09:05 – 10:00 

Coffee break and Poster session 

10:00 – 10:40 

Keynote: Self-supervised and unsupervised learning using diffusion models in medical imaging: from theory to applications (Jong Chul Ye) 

10:40 – 11:10 

Oral Session 2 (2 talks - 15min each) 

  • 010: Bi-directional Synthesis of Pre- and Post-contrast MRI via Guided Feature Disentanglement, Xue, Dewey, Zuo, Han, Carass, Duan, Remedios, Pham, Saidha, Calabresi, Prince (video) 

  • 012: Can segmentation models be trained with fully synthetically generated data? Fernandez, Pinaya, Borges, Tudosiu, Graham , Vercauteren , Cardoso 

11:10 – 11:20 

Sponsor talk  

11:20 – 11:30 

Award and closing ceremony 



Instructions for presenters

For all the papers, please upload 2-minutes poster highlight video and poster PDF before September 10.
2-minutes highlight video
Format: mp4 at 1080p, maximum file size 1 GB, encoded with H.264 (AVCHD) video codec and AAC audio codec. For uploaded video please keep the encoded video bitrate between 4.5Mbps and 6Mbps to balance quality and playback smoothness.
Virtual Poster
Format: pdf, maximum file size 3 MB. We recommend A0 size with at least 32 font size, 72 DPI (3370 x 2384).
In-person presenters only
Physical poster in Portrait format. The maximum poster size for MICCAI 2022 is A0, (i.e. 841 x 1189 mm or 33.1 x 46.8 in) (Width x Height) portrait format. Please adhere to this format.
Oral presenters only
The talk will be 15 min, including Q&A. If you plan to attend SASHIMI virtually, please upload both slides and a 12 min recorded video (.mp4 format) before September 10. If you plan to attend SASHIMI in-person, please email us with the presenter's name and the paper ID/title. We still recommend you upload slides and video to us as a backup.