mlsanityData to 14 Oct 2026

Augmenting medical image classifiers with synthetic data from latent diffusion models

arXiv 2308.12453 ↗First posted 23 Aug 2023cs.CV cs.AI cs.LGcited by 22

On the map: Diffusion models → Medical imaging and segmentation

Abstract (arXiv)

While hundreds of artificial intelligence (AI) algorithms are now approved or cleared by the US Food and Drugs Administration (FDA), many studies have shown inconsistent generalization or latent bias, particularly for underrepresented populations. Some have proposed that generative AI could reduce the need for real data, but its utility in model development remains unclear. Skin disease serves as a useful case study in synthetic image generation due to the diversity of disease appearance, particularly across the protected attribute of skin tone. Here we show that latent diffusion models can scalably generate images of skin disease and that augmenting model training with these data improves performance in data-limited settings. These performance gains saturate at synthetic-to-real image ratios above 10:1 and are substantially smaller than the gains obtained from adding real images. As part of our analysis, we generate and analyze a new dataset of 458,920 synthetic images produced using several generation strategies. Our results suggest that synthetic data could serve as a force-multiplier for model development, but the collection of diverse real-world data remains the most important step to improve medical AI algorithms.

Key facts. Cited by 22 papers in the map (data through 14 Oct 2026). Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models. Builds on 25 of its 45 references that are in the map. First posted on arXiv 23 Aug 2023.

Builds on

25 of 45 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023Diffusion-based Data Augmentation for Skin Disease Classification: Impact Across Original Medical Datasets to Fully Synthetic Images12 Jan
  2. 2022Improving dermatology classifiers across populations using images generated by large diffusion models23 Nov
  3. 2022Generation of Anonymous Chest Radiographs Using Latent Diffusion Models for Training Thoracic Abnormality Classification Systems2 Nov
  4. 2022Adapting Pretrained Vision-Language Foundational Models to Medical Imaging Domains9 Oct

Other references · 21

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  4. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  5. 2021Zero-Shot Text-to-Image Generationcited by 4,288
  6. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,300
  7. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,917
  8. 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,135
  9. 2023Synthetic Data from Diffusion Models Improves ImageNet Classificationcited by 327
  10. 2021Evaluating Deep Neural Networks Trained on Clinical Images in Dermatology with the Fitzpatrick 17k Datasetcited by 185
  11. 2020fastai: A Layered API for Deep Learningcited by 183
  12. 2021Disparities in Dermatology AI: Assessments Using Diverse Clinical Imagescited by 134
9 more
  1. 2019A deep learning system for differential diagnosis of skin diseasescited by 118
  2. 2023Generative models improve fairness of medical classifiers under distribution shiftscited by 90
  3. 2023Differentially Private Diffusion Models Generate Useful Synthetic Imagescited by 86
  4. 2020Overcoming Barriers to Data Sharing with Medical Image Generation: A Comprehensive Evaluationcited by 36
  5. 2021Selective Synthetic Augmentation with HistoGAN for Improved Histopathology Image Classificationcited by 35
  6. 2022Towards Transparency in Dermatology Image Datasets with Skin Tone Annotations by Experts, Crowds, and an Algorithmcited by 35
  7. 2019DermGAN: Synthetic Generation of Clinical Skin Images with Pathologycited by 33
  8. 2021Reducing bias and increasing utility by federated generative modeling of medical images using a centralized adversarycited by 14
  9. 2023Intrinsic Self-Supervision for Data Quality Auditscited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 10; 2025: 5; 2026: 7.

22 papers.

Sort:

By area:

  1. 2024WDM: 3D Wavelet Diffusion Models for High-Resolution Medical Image Synthesiscited by 56
  2. 2024Deep Generative Models for 3D Medical Image Synthesiscited by 14
  3. 2024From Majority to Minority: A Diffusion-based Augmentation for Underrepresented Groups in Skin Lesion Analysiscited by 9
  4. 2025Doctor Approved: Generating Medically Accurate Skin Disease Images through AI-Expert Feedbackcited by 9
  5. 2025Ultrasound Image Generation using Latent Diffusion Modelscited by 7
  6. 2024A Generalist Model for Diverse Text-Guided Medical Image Synthesiscited by 6
  7. 2024S-SYNTH: Knowledge-Based, Synthetic Generation of Skin Imagescited by 6
  8. 2024Trustworthy Text-to-Image Diffusion Models: A Timely and Focused Surveycited by 6
  9. 2024Ctrl-GenAug: Controllable Generative Augmentation for Medical Sequence Classificationcited by 4
  10. 2025LesionGen: A Concept-Guided Diffusion Model for Dermatology Image Synthesiscited by 4
  11. 2024Knowledge-driven AI-generated data for accurate and interpretable breast ultrasound diagnosescited by 2
  12. 2024Bayesian Power Steering: An Effective Approach for Domain Adaptation of Diffusion Modelscited by 1
10 more
  1. 2025GAUDA: Generative Adaptive Uncertainty-guided Diffusion-based Augmentation for Surgical Segmentationcited by 1
  2. 2024Evaluating Utility of Memory Efficient Medical Image Generation: A Study on Lung Nodule Segmentationcited by 0
  3. 2025Automated Measurement of Eczema Severity with Self-Supervised Learningcited by 0
  4. 2026SurrogateSHAP: Training-Free Contributor Attribution for Text-to-Image (T2I) Modelscited by 0
  5. 2026Few-Shot Synthetic Data Generation with Diffusion Models for Downstream Vision Taskscited by 0
  6. 2026Structural MRI Synthesis for Alzheimer's Disease via Conditional Diffusion on Anatomical Maskscited by 0
  7. 2026Controllable Generation of Diverse Dermatological Imagery for Fair and Efficient Malignancy Classificationcited by 0
  8. 2026To do(x) or not to do(x): Medical Image Counterfactuals for Dataset Augmentationcited by 0
  9. 2026Statistical attribute alignment for black-box generative AI via output post-processingcited by 0
  10. 2026Generative AI-Based Data Augmentation for Oral Lesion Classification: The PhotoMOCI Dataset and Benchmarkcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Derm-T2IM: Harnessing Synthetic Skin Lesion Data via Stable Diffusion Models for Enhanced Skin Disease Classification using ViT and CNN
  2. 2026Synthetic Data Generation for Long-Tail Medical Image Classification: A Case Study in Skin Lesions
  3. 2025Prompting Medical Vision-Language Models to Mitigate Diagnosis Bias by Generating Realistic Dermoscopic Images
  4. 2024FairSkin: Fair Diffusion for Skin Disease Image Generation
  5. 2025DermDiff: Generative Diffusion Model for Mitigating Racial Biases in Dermatology Diagnosis
  6. 2025SkinGenBench: Generative Model and Preprocessing Effects for Synthetic Dermoscopic Augmentation in Melanoma Diagnosis
  7. 2025SkinDualGen: Prompt-Driven Diffusion for Simultaneous Image-Mask Generation in Skin Lesions
  8. 2024Cancer-Net SCa-Synth: An Open Access Synthetically Generated 2D Skin Lesion Dataset for Skin Cancer Classification