SASHIMI workshop is jointly organized with SynthRAD2025 Challenge (https://synthrad2025.grand-challenge.org/) as a half-day satellite event (https://conferences.miccai.org/2025/en/workshops.asp) held in conjunction with MICCAI 2025 conference (https://conferences.miccai.org/2025/en/) in Daejeon, Republic of Korea. The workshop will take place on September 23, 2025, the first day of the conference.
| Session 1 Title: SASHIMI |
|---|
| Session Chair: Samuel W. Remedios |
| Time | Duration | Event |
|---|---|---|
| 13:30 - 13:35 | 15 min | Introduction to SASHIMI 2025 (Welcome + Opening remarks) |
| 13:35 - 14:35 | 60 min | Oral presentations |
| FastDTI - Residual Dense Network for DTI | ||
| Jain, A., Ramanarayanan, S., Ram, K., Sivaprakasam, M. | ||
| 3D Super-Resolution for Enhancing Compression Fracture Detection in Thick-Slice CT: Diffusion Models vs GANs | ||
| Kudo, A., Kitamura, Y., Suzuki, Y., Tomiyama, N., Hori, M. | ||
| Conditional Iterative α-(de)Blending Model for CBCT-to-sCT Synthesis: Towards a Deterministic and Simple Process | ||
| Cao, J., Sargeant, C., McWilliam, A., Osorio, E | ||
| Unsupervised MRI Harmonization via Parameter Prediction and Super-Resolved MPMs | ||
| Borges, P., Fernandez, V., Nachev, P., Ourselin, S., Cardoso, J. | ||
| 14:35 - 15:20 | 45 min | Keynote (Invited talk + Q&A) |
| MAISI: A Foundation Model for Accelerated, Anatomy-Aware High-Resolution 3D CT Synthesis | ||
| Can Zhao | ||
| 15:20 - 16:30 | 45 min | Poster session + coffee break |
| SASHIMI posters in addition to the posters corresponding to the above orals: | ||
| MedLoRD: A Medical Low-Resource Diffusion Model for High-Resolution 3D CT Image Synthesis | ||
| Seyfarth, M., Dar, S., Ayx, I., Fink, M., Schönberg, S., Kauczor, H., Engelhardt, S. | ||
| GLFC: Unified Global-Local Feature and Contrast Learning with Mamba-Enhanced UNet for Synthetic CT Generation from CBCT | ||
| Zhou, X., Wu, J., Zhao, H., Chen, L., Zhang, S., Wang, G. | ||
| 2D to 3D MR Image Super-Resolution using Cross-Contrast Guidance | ||
| Zhang, Z., Zhou, Z., Xiang, L., Li, Y., Song, X. | ||
| From Tissue-Mimicking Phantoms to Physics-Based Scans: Synthetic OCT for Few-Shot Foundation Model Training | ||
| Nikoshin, D., Mikhailenko, D., Sovetsky, A., Matveyev, A., Zaitsev, V., Matveev, L. | ||
| Multi-modal Brain MRI Synthesis with nnU-Net: Exploring Segmentation Performance and Cross-Modality Relationships | ||
| Diana-Albelda, C., Longuefosse, A., García-Martín, Á., Bescos, J. | ||
| Unified 3D MRI Representations via Sequence-Invariant Contrastive Learning | ||
| Chalcroft, L., Crinion, J., J. Price, C., Ashburner, J. | ||
| From Lines to Shapes: Geometric-Constrained Segmentation of X-Ray Collimators via Hough Transform | ||
| El-Zein, B., Eckert, D., Fieselmann, A., Syben, C., Ritschl, L., Kappler, S., Stober, S. | ||
| VIOLET: A framework for combined Volumetric Image registration via Optimization and Learning for Efficient image Translation | ||
| Khandelwal, P., Tran Duong, M., Trotman, W., Detre, J., Lee, E., Irwin, D., McMillan, C., Tisdall, D., Das, S., Wolk, D., Yushkevich, P. | ||
| Generation of Controllable and Photorealistic Synthetic Cataract Surgery Images: Blending 3D Models and Real-World Data | ||
| Peter, R., Oberschulte, E., Wu, E., Vaidya, A., Lindemeier, T., Tagliabue, E., Mathis-Ullrich, F. | ||
| Learning mechanistic subtypes of neurodegeneration with a physics-informed variational autoencoder mixture model | ||
| Pinnawala, S., Hartanto, A., Simpson, I., Wijeratne, P. | ||
| Synthesizing Accurate and Realistic T1-weighted Contrast-Enhanced MR Images using Posterior-Mean Rectified Flow | ||
| Brandstötter, B., Kobler, E. | ||
| Clustering-based Stain Augmentation: Templates for Periodic Acid-Schiff Biopsy Images | ||
| Silva, M., Weis, C., Porubsky, S., Leh, S., Weishaupt, H. | ||
| Lesion‐Aware CT-to-MRI Synthesis using a Mask‐Informed Diffusion with Adaptive‐Weighted Loss (MIDAS) | ||
| Mathur, P., Banahan, P., Burns, J., MacMahon, P., Lawlor, A. | ||
| SynthRAD posters: | ||
| Team BreizhCT (SynthRAD Task 1/Rank 3) Why Registration Quality Matters: Enhancing sCT Synthesis with IMPACT-Based Registration |
||
| Boussot V., Hemon C., Nunes J., Dillenseger J. | ||
| Team MixCT (SynthRAD Task 1/Rank 4) GANeXt: A Fully ConvNeXt-Enhanced Generative Adversarial Network for MRI- and CBCT-to-CT Synthesis |
||
| Mei S., Xia Y., Fan F. | ||
| Team QWER (SynthRAD Task 1/Rank 5) A MR-to-CT Synthesis Method for SynthRAD2025 Challenge |
||
| Han B., Tan Z. | ||
| Team DeepSyn (SynthRAD Task 1/Rank 6) Research on MR-Based Synthetic CT using Deep Learning |
||
| Hou X., Ji J., Yang H. | ||
| Team Faking_it (SynthRAD Task 1/Rank 7) VS-DDPM: Efficient Low-Cost Diffusion Model for Medical Modality Translation |
||
| Moradi N., Luijten G., Puladi B., Kleesiek J., Alves V., Egger J., Ferreira A | ||
| Team MEVIS (SynthRAD Task 1/Rank 8) Patch-based 3D Hybrid UNet for MRI to CT Synthesis |
||
| Mensing D., Heldmann S | ||
| Team HiLab (SynthRAD Task 1/Rank 9) Method Description for SynthRAD2025 Task 1: MRI-to-CT Translation |
||
| Zhou X., Zhao K., Zhang S., Wang G. | ||
| Team sk (SynthRAD Task 1/Rank 10) Synthetic CT Generation from MRI and CBCT Using U-Net with Pretrained ConvNeXt Encoder |
||
| Kondo S., Kasai S. | ||
| Team BreizhCT (SynthRAD Task 2/Rank 3) Why Registration Quality Matters: Enhancing sCT Synthesis with IMPACT-Based Registration |
||
| Boussot V., Hemon C., Nunes J., Dillenseger J. | ||
| Team Et (SynthRAD Task 2/Rank 4) Multi-Slice UNet++ for CBCT-CT Synthesis |
||
| Zhang Z., Cai J. | ||
| Team RicardoBrioso (SynthRAD Task 2/Rank 5) ATUNet: Attention-gated Tiny U-Net for CBCT-to-CT Synthesis using a Hybrid Perceptual Loss |
||
| Brioso R.C., Dei D., Lambri N., Mancosu P., Scorsetti M., Loiacono D. | ||
| Team SynthRAD Shu (SynthRAD Task 2/Rank 6) Two-stage Full-parameter Fine-tuning 3D GAN for CBCT-CT Synthesis in SynthRAD2025 |
||
| Liu J., Wang Q., Tang Y. | ||
| Team HiLab (SynthRAD Task 2/Rank 7) Method Description for SynthRAD2025 Task 2: CBCT-to-CT Translation |
||
| Zhou X., Zhao K., Zhang S., Wang G. | ||
| Team Faking_it (SynthRAD Task 2/Rank 8) VS-DDPM: Efficient Low-Cost Diffusion Model for Medical Modality Translation |
||
| Moradi N., Luijten G., Puladi B., Kleesiek J., Alves V., Egger J., Ferreira A. | ||
| Team KoalAI (SynthRAD Task 2/Rank 9) Ensembled ResUnet with Masked Anatomical Perception Loss |
||
| Xin B., Sun Z., Min H., Belous G., Dowling J. | ||
| Team abdullahshazli18 (SynthRAD Task 2/Rank 10) Anatomically Constrained DynUNet for Task 2: CBCT-to-sCT Generation |
||
| Shazly A., Rezk A.M., Al-Fahik A., Kim D., Ryu K., Al-Masni M.A. |
| Session 2 Title: SynthRAD2025 |
|---|
| Session Chairs: Adrian Thummerer, Viktor Rogowski |
| Time | Duration | Event |
|---|---|---|
| 16:30 - 16:42 | 12 min | Introduction to SynthRAD2025 |
| 16:47 - 17:00 | 13 min | 2nd place task 1 & 2: Team ImagePasNet |
| Deep Learning-Based Cross-Anatomy CT Synthesis Using Adapted nnResU-Net with Anatomical Feature Prioritized Loss | ||
| Sequeiro Gonzalez J., Longuefosse A., Diaz Benito M., Martin A.G., Baldacci F. | ||
| 17:01 - 17:14 | 13 min | Winner task 1: Team KoalAI |
| Ensembled ResUnet with Masked Anatomical Perception LossEnsembled ResUnet with Masked Anatomical Perception Loss | ||
| Xin B., Sun Z., Min H., Belous G., Dowling J. | ||
| 17:15 - 17:27 | 12 min | Winner task 2: Team MixCT |
| GANeXt: A Fully ConvNeXt-Enhanced Generative Adversarial Network for MRI- and CBCT-to-CT Synthesis | ||
| Mei S., Xia Y., Fan F.Mei S., Xia Y., Fan F. | ||
| 17:27 - 18:00 | 33 min | Closing |