Research map / Image generation / Text-to-image diffusion
MedSegFactory: Text-Guided Generation of Medical Image-Mask Pairs
Jiawei Mao, Yuhan Wang, Yucheng Tang, Daguang Xu, Kang Wang, Yang Yang, Zongwei Zhou, Yuyin Zhou
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 |
|---|---|---|---|
| add Joint Cross-Attention between image and mask streams (ablation) | MedSegFactory · average over 6 datasets | average DSC (%): 61.58 -> 76.57 | Improved |
| add mask-to-image branch in JCA (ablation) | MedSegFactory · average over 6 datasets | average FID: 151.71 -> 138.03 | Improved |
| add image-to-mask branch in JCA (ablation) | MedSegFactory · average over 6 datasets | average DSC (%): 69.83 -> 76.57 | Improved |
| augment nnUNet training with MedSegFactory synthetic image-mask pairs | nnUNet · BUSI | DSC (%): 78.77 -> 83.38 | Improved |
| augment nnUNet training with MedSegFactory synthetic data | nnUNet · LiTS17 | DSC (%): 59.77 -> 62.37 | Improved |
| augment nnUNet with MedSegFactory data versus ControlNet data on AMOS and CVC-ClinicDB | nnUNet · AMOS | DSC (%): 77.16 -> 76.76 (ControlNet vs MedSegFactory) | Mixed |
| use MedSegFactory instead of DiffBoost for abdominal CT generation | image generation · AMOS, KiTS19 | FID: 73.61 -> 97.94 (AMOS) | Hurt |
| generate images without segmentation targets | MedSegFactory · target-free generation (qualitative) | boundary quality | Hurt |
Closest papers by meaning
- Noise-Consistent Siamese-Diffusion for Medical Image Synthesis and Segmentation (CVPR 2025 · similarity 0.86)
- MedFlowSeg: Flow Matching for Medical Image Segmentation with Frequency-Aware Attention (NeurIPS 2026 · similarity 0.86)
- MAPSeg: Unified Unsupervised Domain Adaptation for Heterogeneous Medical Image Segmentation Based on 3D Masked Autoencoding and Pseudo-Labeling (CVPR 2024 · similarity 0.85)
- Unleashing the Potential of SAM for Medical Adaptation via Hierarchical Decoding (CVPR 2024 · similarity 0.85)
- Interactive Medical Image Segmentation: A Benchmark Dataset and Baseline (CVPR 2025 · similarity 0.84)
- MedSegDiff-V2: Diffusion-Based Medical Image Segmentation with Transformer (AAAI 2024 · 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.
