mlsanityData to 14 Oct 2026

MedSymmFlow: Bridging Generative Modeling and Classification in Medical Imaging through Symmetrical Flow Matching

arXiv 2507.19098 ↗First posted 25 Jul 2025cs.CV cs.AIcited by 2

On the map: Diffusion models → Medical imaging and segmentation

Abstract (arXiv)

Reliable medical image classification requires accurate predictions and well-calibrated uncertainty estimates, especially in high-stakes clinical settings. This work presents MedSymmFlow, a generative-discriminative hybrid model built on Symmetrical Flow Matching, designed to unify classification, generation, and uncertainty quantification in medical imaging. MedSymmFlow leverages a latent-space formulation that scales to high-resolution inputs and introduces a semantic mask conditioning mechanism to enhance diagnostic relevance. Unlike standard discriminative models, it naturally estimates uncertainty through its generative sampling process. The model is evaluated on four MedMNIST datasets, covering a range of modalities and pathologies. The results show that MedSymmFlow matches or exceeds the performance of established baselines in classification accuracy and AUC, while also delivering reliable uncertainty estimates validated by performance improvements under selective prediction.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models. Builds on 20 of its 33 references that are in the map. First posted on arXiv 25 Jul 2025.

Builds on

20 of 33 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025Conditional Diffusion Models are Medical Image Classifiers that Provide Explainability and Uncertainty for Free6 Feb
  2. 2024Data Augmentation for Surgical Scene Segmentation with Anatomy-Aware Diffusion Models10 Oct
  3. 2024MAISI: Medical AI for Synthetic Imaging13 Sep
  4. 2023Unsupervised 3D out-of-distribution detection with latent diffusion models7 Jul
  5. 2023MedSegDiff-V2: Diffusion based Medical Image Segmentation with Transformer19 Jan
  6. 2023Diffusion-based Data Augmentation for Skin Disease Classification: Impact Across Original Medical Datasets to Fully Synthetic Images12 Jan

Other references · 14

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  3. 2022Flow Matching for Generative Modelingcited by 5,702
  4. 2018Neural Ordinary Differential Equationscited by 3,779
  5. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 972
  6. 2021MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classificationcited by 690
  7. 2018FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Modelscited by 594
  8. 2021Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysiscited by 422
  9. 2023Your Diffusion Model is Secretly a Zero-Shot Classifiercited by 306
  10. 2018Auto-Keras: An Efficient Neural Architecture Search Systemcited by 171
  11. 2023MedViT: A Robust Vision Transformer for Generalized Medical Image Classificationcited by 89
  12. 2020Neural Networks with Recurrent Generative Feedbackcited by 23
2 more
  1. 2025Symmetrical Flow Matching: Unified Image Generation, Segmentation, and Classification with Score-Based Generative Modelscited by 8
  2. 2024EffiSegNet: Gastrointestinal Polyp Segmentation through a Pre-Trained EfficientNet-based Network with a Simplified Decodercited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Representation-driven Endoscopic Visual Embedding Alignment for Latent Generationcited by 0
  2. 2026SymmAdapt: Symmetrical Flow Matching for Source-Free Domain Adaptation in Medical Image Segmentationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025LatentFM: A Latent Flow Matching Approach for Generative Medical Image Segmentation
  2. 2026Flow Matching with Optimized Subclass Priors for Medical Image Augmentation
  3. 2025Aleatoric Uncertainty Medical Image Segmentation Estimation via Flow Matching
  4. 2025Learning Patient-Specific Disease Dynamics with Latent Flow Matching for Longitudinal Imaging Generation
  5. 2026Few-Shot Distribution-Aligned Flow Matching for Data Synthesis in Medical Image Segmentation
  6. 2026Uncertainty-Aware Distribution-to-Distribution Flow Matching for Scientific Imaging
  7. 2026The Learnability Gap in Medical Latent Diffusion
  8. 2026MedFlowSeg: Flow Matching for Medical Image Segmentation with Frequency-Aware Attention