Medical image generation using deep learning: a systematic review
Applied Intelligence · 2026-11 · 56 (17)
DOI: 10.1007/s10489-026-07572-9 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift Applied Intelligence, online erschienen am 19.11.2026. Verfasst von Mengdi Gao und 7 weiteren Autor:innen. Die Arbeit nennt 236 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Beijing University of Technology High-End Foreign Experts Recruitment Program.
Abstract (Originalsprache)
Medical image analysis is central to modern healthcare. However, the development of robust deep learning models is frequently hindered by a scarcity of large-scale, diverse, and well-annotated datasets. Medical image generation has therefore emerged as an effective solution, enabling data augmentation, domain adaptation, and the synthesis of realistic pathological cases. This survey first outlines publicly available medical imaging datasets widely used in generation tasks and discusses key quantitative metrics for assessing the fidelity and clinical utility of synthesized images. Next, it presents a comprehensive review of state-of-the-art medical image generation techniques, systematically categorizing them into four principal frameworks: Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), Diffusion Models, and hybrid approaches integrating these methodologies. It then provides an in-depth analysis of the unique characteristics, advantages, and limitations of each category, alongside their significant contributions to downstream medical applications. Finally, it identifies persistent challenges and proposes promising research directions to guide future advancements in medical image generation.
Förderung
- National Natural Science Foundation of China
- Beijing University of Technology High-End Foreign Experts Recruitment Program
Zitieren
APA
Gao, M., Wang, J., Qu, Y., Zhao, F., Ding, L., Chen, Z., Yang, L., & Jiang, H. (2026). Medical image generation using deep learning: a systematic review. Applied Intelligence, 56(17). https://doi.org/10.1007/s10489-026-07572-9
Vancouver
Gao M, Wang J, Qu Y, Zhao F, Ding L, Chen Z, et al. Medical image generation using deep learning: a systematic review. Applied Intelligence. 2026;56(17). doi:10.1007/s10489-026-07572-9
BibTeX
@article{Gao2026,
author = {Gao, Mengdi and Wang, Jiarui and Qu, Yanting and Zhao, Fang and Ding, Langlang and Chen, Zhuo and Yang, Lin and Jiang, Hongyang},
title = {Medical image generation using deep learning: a systematic review},
journal = {Applied Intelligence},
year = {2026},
volume = {56},
number = {17},
doi = {10.1007/s10489-026-07572-9}
}
EUVIMED