DSIN: 动态系统赋能的隐式神经表示在医学图像压缩中的应用
DSIN: dynamic system-empowered implicit neural representation for medical image compression
DSIN 是一种用于医学图像压缩的新框架,通过图像特定的隐式神经表示和基于动态系统的压缩模块,在保持重建质量的同时实现了高达 183 倍的压缩比。该方法在多种医学影像数据集(包括 CT、MRI、PET、CTA 和病理图像)上的测试中表现出色,优于现有方法。
- Article
- Open access
- Published: 03 November 2026
- Hanxiao Song1,
- Huaxian Shi1,
- Ying Li1,
- Hengtao Guo2,
- Haoran Wang3,
- Jianhua Ma1,4,
- Fenglei Fan3 &
- …
- Dong Zeng ORCID: orcid.org/0000-0001-6015-50101
Communications Engineering (2026) Cite this article
We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.
Abstract
Multidimensional/Ultra-high-resolution medical imaging, e.g., Photon-counting computed tomography, multiparametric Magnetic resonance imaging (MRI), whole-slide digital pathology drives diagnostic precision but generates massive datasets that strain clinical infrastructure. Current compression methods either sacrifice subvisual diagnostic features to prioritize human perception or lack generalizability across modalities and pathologies. Here, we introduce Dynamic system-empowered implicit neural representation (DSIN), an image compression framework using image-specific implicit neural representations, where a stacked dynamic-system-based compression module further reduces model size while preserving reconstruction quality. When tested across five large-scale clinical datasets (including Computed tomography (CT), MRI, Positron emission tomography (PET), Computed tomography angiography (CTA), and histopathology images), DSIN achieves compression ratios up to 183-fold while maintaining high reconstruction quality, outperforming established and implicit neural representation methods in downstream tasks such as segmentation and classification. Quantitative evaluations reveal superior performance in preserving reconstruction quality, organ boundary fidelity, and metabolic activity delineation in low-dose PET imaging. DSIN provides an effective compression framework for the medical imaging modalities evaluated in this study.
Funding
This research was supported by the National Key R&D Program of China (Grant No. 2024YFA1012003 to J.M.) and the National Natural Science Foundation of China (Grant No. U21A6005 to D.Z.).
Author information
Authors and Affiliations
School of Biomedical Engineering, Southern Medical University, Guangzhou, China
Hanxiao Song, Huaxian Shi, Ying Li, Jianhua Ma & Dong Zeng
Independent Researcher, Seattle, WA, USA
Hengtao Guo
Department of Data Science, City University of Hong Kong, Hong Kong, China
Haoran Wang & Fenglei Fan
School of Life Science and Technology, Xi’an Jiaotong University, Xi’an, China
Jianhua Ma
Authors
- Hanxiao Song
- Huaxian Shi
- Ying Li
- Hengtao Guo
- Haoran Wang
- Jianhua Ma
- Fenglei Fan
- Dong Zeng
Corresponding authors
Correspondence to Jianhua Ma, Fenglei Fan or Dong Zeng.
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Song, H., Shi, H., Li, Y. et al. DSIN: dynamic system-empowered implicit neural representation for medical image compression. Commun Eng (2026). https://doi.org/10.1038/s44172-026-00812-3
Received: 22 December 2025
Accepted: 01 October 2026
Published: 03 November 2026
DOI: https://doi.org/10.1038/s44172-026-00812-3
来源:Nature 健康护理主题 · nature.com