Research map / Image generation / Autoregressive and latent generation
Foundation VAE for CT Reconstruction, Augmentation, and Generation
Qi Chen, Shuhan Ding, Yu Gu, Nan Liu, Jiang Bian, Alan Yuille, Zongwei Zhou, Jingjing Fu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Train segmenter on frozen Foundation VAE reconstructions instead of real CT | nnU-Net (WAN2.1) · MSD Task06 Lung | NSD: 70.2 -> 75.3 | Improved |
| Use CVVAE reconstructions for segmentation training instead of real CT | nnU-Net · MSD Task07 Pancreas (tumor) | NSD2: 39.8 -> 34.5 | Hurt |
| Replace Foundation VAE with MedVAE in latent diffusion (ablation) | conditional latent diffusion · CT-RATE normal split | FID: 11.28 -> 2.19 | Improved |
| Use Foundation VAE instead of MedVAE (ablation) | conditional latent diffusion · CT-RATE normal split | CT-CLIP: 20.76 -> 59.35 | Improved |
| Add a single disease mask and single-finding prompt to organ-mask conditioning (ablation) | conditional latent diffusion · CT-RATE disease split | CT-CLIP Avg: 43.40 -> 46.14 | Improved |
| Use multi-disease masks with full report prompts (ablation) | conditional latent diffusion · CT-RATE disease split | CT-CLIP Avg: CT-CLIP 46.14 -> 51.49; FID 3.76 -> 4.35 | Mixed |
| Add disease-mask-guided synthetic CT to classifier training | multi-label classifier · CT-RATE 18-label classification | mean AUC: 67.95 -> 70.71 | Improved |
| Condition generation on organ masks (MAISI-style masks) with text and disease mask | Ours vs MAISI · Organ-mask adherence | vessel Dice: 13.50 -> 63.38 | Improved |
| Use CT-specific MedVAE for reconstruction on MSD data | MedVAE · MSD Task07 Pancreas | PSNR: PSNR 36.00 vs IVVAE 40.43 | Hurt |
Closest papers by meaning
- SSC-VAE: Structured Sparse Coding Based Variational Autoencoder for Detail Preserved Image Reconstruction (AAAI 2025 · similarity 0.87)
- Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models (CVPR 2025 · similarity 0.85)
- Better Tokens for Better 3D: Advancing Vision-Language Modeling in 3D Medical Imaging (NeurIPS 2025 · similarity 0.84)
- CV-VAE: A Compatible Video VAE for Latent Generative Video Models (NeurIPS 2024 · similarity 0.84)
- VCT: Training Consistency Models with Variational Noise Coupling (ICML 2025 · similarity 0.84)
- Improved Video VAE for Latent Video Diffusion Model (CVPR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
