SASHIMI 2021: Simulation and Synthesis in Medical Imaging


SASHIMI 2021 will be held on September 27. The program includes a keynote talk, six oral presentations, and two poster sessions preceded by poster highlights. Times specified correspond to the UTC time zone.

  •  First half-day session

09:00 – 09:10 Introduction (Live)
09:10 – 10:00 Oral Session I (Pre-recorded talks with live Q&A)
09:10 – 09:30 SequenceGAN: Generating Fundus Fluorescence Angiography Sequences from Structure Fundus Image (paper)
Wanyue Li, Wen Kong, Yi He, Jing Wang, Guohua Deng, Yiwei Chen, Guohua Shi
09:30 – 09:50 The role of MRI physics in brain segmentation CNNs: achieving acquisition invariance and instructive uncertainties (paper | code)
Pedro Borges, Richard Shaw, Thomas Varsavsky, Kerstin Klaser, David Thomas, Ivana Drobnjak, Sebastien Ourselin, M. Jorge Cardoso
9:50 – 10:00 Poster Highlights I (Pre-recorded videos)
10:00 – 10:25 Break   
10:25 – 11:25 Poster Session I (Live on Spatial Chat)
  Learning-based Template Synthesis For Groupwise Image Registration (paper)
Ziyi He, Albert C. S. Chung
  Super-resolution by Latent Space Exploration: Training with Poorly- aligned Clinical and Micro CT Image Dataset (paper)
Tong Zheng, Hirohisa Oda, Yuichiro Hayashi, Shota Nakamura, Masahiro Oda, Kensaku Mori
  A Glimpse into the Future: Disease Progression Simulation for Breast Cancer in Mammograms (paper | video)
Ibrahim Jubran, Moshe Raboh, Shaked Perek, David Gruen, Efrat Hexter
  X-ray synthesis based on triangular mesh models using GPU- accelerated ray tracing for multi-modal breast image registration (paper | video | poster | code)
Johannes Maul, Sarah Said, Nicole Ruiter, Torsten Hopp
11:25 – 11:40 Break  
11:40 – 12:25 Oral Session II (Pre-recorded talks with live Q&A)
11:40 – 11:45 Surprise! 
11:45 – 11:55  A word by Hyperfine, award sponsor   
11:55 – 12:15 Cerebral Blood Volume Prediction based on Multi-modality Magnetic Resonance Imaging (paper)
Yongsheng Pan, Jingyu Huang, Bao Wang, Peng Zhao, Yingchao Liu, Yong Xia
12:15 – 12:35 Detail matters: high-frequency content for realistic synthetic brain MRI generation (paper | video)
Filip Rusak, Rodrigo santa Cruz, Elliot Smith, Jurgen Fripp, Clinton Fookes, Pierrick Bourgeat, Andrew P. Bradley
  • Second half-day session

14:00 – 15:00 Keynote Talk (Pre-recorded talk with live Q&A)    
Deep Optics: Learning Cameras and Optical Computing Systems Gordon Wetzstein
15:00 – 15:10 Poster Highlights II (Pre-recorded videos) 
15:10 – 15:25 Break   
15:25 – 16:25 Poster Session II (Live on Spatial Chat)
  Frozen-to-Paraffin: Categorization of Histological Frozen Sections by the Aid of Paraffin Sections and Generative Adversarial Networks (paper)
Michael Gadermayr, Maximilian Ernst Tschuchnig, Lea Maria Stangassinger, Christina Kreutzer, Sebastien Couillard-Despres, Gertie Janneke Oostingh, Anton Hittmair
  Transfer Learning in Optical Microscopy (paper)
Martin Kozlovský, David Wiesner, David Svoboda
  Cine-MRI simulation to evaluate tumor tracking (paper | code)
José D. Tascón-Vidarte, Isak Wahlstedt, Julien Jomier, Kenny Erleben, Ivan R. Vogelius, Sune Darkner
GAN-based synthetic FDG PET images from T1 brain MRI can serve to improve performance of deep unsupervised anomaly detection models (paper)
Daria Zotova, Julien Jung, Carole Lartizien
16:25 – 16:40 Break  
16:40 – 17:30 Oral Session III (Pre-recorded talks with live Q&A)
16:40 – 16:50 A word by Nvidia, award sponsor 
16:50 – 17:10 Synth-by-Reg (SbR): Contrastive learning for synthesis-based registration of paired images (paper | code)
Adrià Casamitjana, Matteo Mancini, Juan Eugenio Iglesias
17:10 – 17:30 Joint Image and Label Self-Super-Resolution (paper | code)
Samuel Remedios, Shuo Han, Blake Dewey, Dzung Pham, Jerry Prince, Aaron Carass
17:30 – 17:40 Best paper awards & Closing remarks (Live)   

Updated on 24/09/2021

 Best paper awards

We are very pleased to announce that

Adrià Casamitjana, Matteo Mancini and Juan Eugenio Iglesias

received the best paper award, a RTX 3090 GPU graciously sponsored by Nvidia, for their paper entitled

"Synth-by-Reg (SbR): Contrastive learning for synthesis-based registration of paired images"

and that

Johannes Maul, Sarah Said, Nicole Ruiter and Torsten Hopp

received the reproducibility award$200 graciously sponsored by Hyperfine, for their paper entitled

"X-ray synthesis based on triangular mesh models using GPU-accelerated ray tracing for multi-modal breast image registration",

which rewards the best paper with publicly available, well documented, and easy to use source code.


NVIDIA Logo H ForScreen ForLightBG