Verdiff-Net: A Conditional Diffusion Framework for Spinal Medical Image Segmentation

Zhang, Zhiqing; Liu, Tianyong; Fan, Guojia; Pu, Yao; Li, Bin; Chen, Xingyu; Feng, Qianjin; Zhou, Shoujun

doi:10.3390/bioengineering11101031

Open AccessArticle

Verdiff-Net: A Conditional Diffusion Framework for Spinal Medical Image Segmentation

by

Zhiqing Zhang

^1,2,

Tianyong Liu

³,

Guojia Fan

⁴,

Yao Pu

⁵

,

Bin Li

¹

,

Xingyu Chen

¹,

Qianjin Feng

^6,* and

Shoujun Zhou

^1,*

¹

Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China

²

University of Chinese Academy of Sciences, Beijing 100049, China

³

Institute of Unconventional Oil & Gas Research, Northeast Petroleum University, Daqing 163318, China

⁴

College of Information Science and Engineering, Northeastern University, Shenyang 110819, China

⁵

Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Kowloon, Hong Kong SAR, China

⁶

School of Biomedical Engineering, Southern Medical University, Guangzhou 510515, China

^*

Authors to whom correspondence should be addressed.

Bioengineering 2024, 11(10), 1031; https://doi.org/10.3390/bioengineering11101031

Submission received: 3 September 2024 / Revised: 26 September 2024 / Accepted: 10 October 2024 / Published: 15 October 2024

(This article belongs to the Section Biosignal Processing)

Download

Browse Figures

Versions Notes

Abstract

Spinal medical image segmentation is critical for diagnosing and treating spinal disorders. However, ambiguity in anatomical boundaries and interfering factors in medical images often cause segmentation errors. Current deep learning models cannot fully capture the intrinsic data properties, leading to unstable feature spaces. To tackle the above problems, we propose Verdiff-Net, a novel diffusion-based segmentation framework designed to improve segmentation accuracy and stability by learning the underlying data distribution. Verdiff-Net integrates a multi-scale fusion module (MSFM) for fine feature extraction and a noise semantic adapter (NSA) to refine segmentation masks. Validated across four multi-modality spinal datasets, Verdiff-Net achieves a high Dice coefficient of 93%, demonstrating its potential for clinical applications in precision spinal surgery.

Keywords: spinal segmentation; diffusion model; multi-modality

Graphical Abstract

Share and Cite

MDPI and ACS Style

Zhang, Z.; Liu, T.; Fan, G.; Pu, Y.; Li, B.; Chen, X.; Feng, Q.; Zhou, S. Verdiff-Net: A Conditional Diffusion Framework for Spinal Medical Image Segmentation. Bioengineering 2024, 11, 1031. https://doi.org/10.3390/bioengineering11101031

AMA Style

Zhang Z, Liu T, Fan G, Pu Y, Li B, Chen X, Feng Q, Zhou S. Verdiff-Net: A Conditional Diffusion Framework for Spinal Medical Image Segmentation. Bioengineering. 2024; 11(10):1031. https://doi.org/10.3390/bioengineering11101031

Chicago/Turabian Style

Zhang, Zhiqing, Tianyong Liu, Guojia Fan, Yao Pu, Bin Li, Xingyu Chen, Qianjin Feng, and Shoujun Zhou. 2024. "Verdiff-Net: A Conditional Diffusion Framework for Spinal Medical Image Segmentation" Bioengineering 11, no. 10: 1031. https://doi.org/10.3390/bioengineering11101031

APA Style

Zhang, Z., Liu, T., Fan, G., Pu, Y., Li, B., Chen, X., Feng, Q., & Zhou, S. (2024). Verdiff-Net: A Conditional Diffusion Framework for Spinal Medical Image Segmentation. Bioengineering, 11(10), 1031. https://doi.org/10.3390/bioengineering11101031

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Verdiff-Net: A Conditional Diffusion Framework for Spinal Medical Image Segmentation

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI