Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet

Yao Pu; Yifan Zhu; Jingguo Qu; Xiang Wang; Wen Li; Mayang Zhao; Peixin Yu; Zihan Li; Xinzhi Teng; Xinyu Zhang; Jiang Zhang; Tao Peng; Jing Cai; Ge Ren

doi:10.1007/978-3-032-06624-4_16

Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet

Yao Pu
, Yifan Zhu
, Jingguo Qu
, Xiang Wang
, Wen Li
, Mayang Zhao
, Peixin Yu
, Zihan Li
, Xinzhi Teng
, Xinyu Zhang
, Jiang Zhang
, Tao Peng
, Jing Cai
, Ge Ren

Department of Health Technology and Informatics

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

Abstract

Breast cancer poses a serious risk to women’s health, making early detection critical. Dynamic Contrast-Enhanced MRI (DCE-MRI) aids diagnosis by using a contrast agent to highlight tissue details, but its cost and health risks limit accessibility. This study proposes a conditional Generative Adversarial Network (cGAN) to enhance MRI images, incorporating a novel generator (VMKANet, based on Vision Mamba and Kolmogorov-Arnold Network) and a discriminator (CAViT, using Vision Transformer with Cross Attention). To our knowledge, this is the first method to generate Breast DCE-MRI images from single-modality MRI data using Mamba. Our approach shows significant improvements in image quality, validated by strong Spearman correlation (R), high structural similarity index (SSIM), enhanced mutual information (MI), reduced normalized root mean square error (NRMSE), and lower symmetric mean absolute percent error (SMAPE). Results demonstrate that VMKANet and CAViT effectively enhance breast MRI images, promising safer and more affordable diagnostic capabilities.

Original language	English
Title of host publication	Computational Mathematics Modeling in Cancer Analysis - 4th International Workshop, CMMCA 2025, Held in Conjunction with MICCAI 2025, Proceedings
Editors	Chao Li, Wenjian Qin, Jia Wu, Nazar Zaki
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	150-159
Number of pages	10
Volume	16178
ISBN (Print)	9783032066237
DOIs	https://doi.org/10.1007/978-3-032-06624-4_16
Publication status	Published - 25 Sept 2025
Event	4th International Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2025, Held in Conjunction with the 28th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2025 - Daejeon, Korea, Republic of Duration: 27 Sept 2025 → 27 Sept 2025

Publication series

Name	Lecture Notes in Computer Science
Volume	16178 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	4th International Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2025, Held in Conjunction with the 28th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2025
Country/Territory	Korea, Republic of
City	Daejeon
Period	27/09/25 → 27/09/25

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Attention
Breast Cancer
cGAN MRI Enhancement
KAN
Mamba

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

More information

10.1007/978-3-032-06624-4_16

Cite this

Pu, Y., Zhu, Y., Qu, J., Wang, X., Li, W., Zhao, M., Yu, P., Li, Z., Teng, X., Zhang, X., Zhang, J., Peng, T., Cai, J., & Ren, G. (2025). Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet. In C. Li, W. Qin, J. Wu, & N. Zaki (Eds.), Computational Mathematics Modeling in Cancer Analysis - 4th International Workshop, CMMCA 2025, Held in Conjunction with MICCAI 2025, Proceedings (Vol. 16178, pp. 150-159). (Lecture Notes in Computer Science; Vol. 16178 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-032-06624-4_16

Pu, Yao ; Zhu, Yifan ; Qu, Jingguo et al. / Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet. Computational Mathematics Modeling in Cancer Analysis - 4th International Workshop, CMMCA 2025, Held in Conjunction with MICCAI 2025, Proceedings. editor / Chao Li ; Wenjian Qin ; Jia Wu ; Nazar Zaki. Vol. 16178 Springer Science and Business Media Deutschland GmbH, 2025. pp. 150-159 (Lecture Notes in Computer Science).

@inproceedings{26b8e42fe2b04d20a263d18f1946689c,

title = "Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet",

abstract = "Breast cancer poses a serious risk to women{\textquoteright}s health, making early detection critical. Dynamic Contrast-Enhanced MRI (DCE-MRI) aids diagnosis by using a contrast agent to highlight tissue details, but its cost and health risks limit accessibility. This study proposes a conditional Generative Adversarial Network (cGAN) to enhance MRI images, incorporating a novel generator (VMKANet, based on Vision Mamba and Kolmogorov-Arnold Network) and a discriminator (CAViT, using Vision Transformer with Cross Attention). To our knowledge, this is the first method to generate Breast DCE-MRI images from single-modality MRI data using Mamba. Our approach shows significant improvements in image quality, validated by strong Spearman correlation (R), high structural similarity index (SSIM), enhanced mutual information (MI), reduced normalized root mean square error (NRMSE), and lower symmetric mean absolute percent error (SMAPE). Results demonstrate that VMKANet and CAViT effectively enhance breast MRI images, promising safer and more affordable diagnostic capabilities.",

keywords = "Attention, Breast Cancer, cGAN MRI Enhancement, KAN, Mamba",

author = "Yao Pu and Yifan Zhu and Jingguo Qu and Xiang Wang and Wen Li and Mayang Zhao and Peixin Yu and Zihan Li and Xinzhi Teng and Xinyu Zhang and Jiang Zhang and Tao Peng and Jing Cai and Ge Ren",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.; 4th International Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2025, Held in Conjunction with the 28th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2025 ; Conference date: 27-09-2025 Through 27-09-2025",

year = "2025",

month = sep,

day = "25",

doi = "10.1007/978-3-032-06624-4\_16",

language = "English",

isbn = "9783032066237",

volume = "16178",

series = "Lecture Notes in Computer Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "150--159",

editor = "Chao Li and Wenjian Qin and Jia Wu and Nazar Zaki",

booktitle = "Computational Mathematics Modeling in Cancer Analysis - 4th International Workshop, CMMCA 2025, Held in Conjunction with MICCAI 2025, Proceedings",

address = "Germany",

}

Pu, Y, Zhu, Y, Qu, J, Wang, X, Li, W, Zhao, M, Yu, P, Li, Z, Teng, X, Zhang, X, Zhang, J, Peng, T, Cai, J & Ren, G 2025, Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet. in C Li, W Qin, J Wu & N Zaki (eds), Computational Mathematics Modeling in Cancer Analysis - 4th International Workshop, CMMCA 2025, Held in Conjunction with MICCAI 2025, Proceedings. vol. 16178, Lecture Notes in Computer Science, vol. 16178 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 150-159, 4th International Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2025, Held in Conjunction with the 28th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2025, Daejeon, Korea, Republic of, 27/09/25. https://doi.org/10.1007/978-3-032-06624-4_16

Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet. / Pu, Yao; Zhu, Yifan; Qu, Jingguo et al.
Computational Mathematics Modeling in Cancer Analysis - 4th International Workshop, CMMCA 2025, Held in Conjunction with MICCAI 2025, Proceedings. ed. / Chao Li; Wenjian Qin; Jia Wu; Nazar Zaki. Vol. 16178 Springer Science and Business Media Deutschland GmbH, 2025. p. 150-159 (Lecture Notes in Computer Science; Vol. 16178 LNCS).

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet

AU - Pu, Yao

AU - Zhu, Yifan

AU - Qu, Jingguo

AU - Wang, Xiang

AU - Li, Wen

AU - Zhao, Mayang

AU - Yu, Peixin

AU - Li, Zihan

AU - Teng, Xinzhi

AU - Zhang, Xinyu

AU - Zhang, Jiang

AU - Peng, Tao

AU - Cai, Jing

AU - Ren, Ge

PY - 2025/9/25

Y1 - 2025/9/25

N2 - Breast cancer poses a serious risk to women’s health, making early detection critical. Dynamic Contrast-Enhanced MRI (DCE-MRI) aids diagnosis by using a contrast agent to highlight tissue details, but its cost and health risks limit accessibility. This study proposes a conditional Generative Adversarial Network (cGAN) to enhance MRI images, incorporating a novel generator (VMKANet, based on Vision Mamba and Kolmogorov-Arnold Network) and a discriminator (CAViT, using Vision Transformer with Cross Attention). To our knowledge, this is the first method to generate Breast DCE-MRI images from single-modality MRI data using Mamba. Our approach shows significant improvements in image quality, validated by strong Spearman correlation (R), high structural similarity index (SSIM), enhanced mutual information (MI), reduced normalized root mean square error (NRMSE), and lower symmetric mean absolute percent error (SMAPE). Results demonstrate that VMKANet and CAViT effectively enhance breast MRI images, promising safer and more affordable diagnostic capabilities.

AB - Breast cancer poses a serious risk to women’s health, making early detection critical. Dynamic Contrast-Enhanced MRI (DCE-MRI) aids diagnosis by using a contrast agent to highlight tissue details, but its cost and health risks limit accessibility. This study proposes a conditional Generative Adversarial Network (cGAN) to enhance MRI images, incorporating a novel generator (VMKANet, based on Vision Mamba and Kolmogorov-Arnold Network) and a discriminator (CAViT, using Vision Transformer with Cross Attention). To our knowledge, this is the first method to generate Breast DCE-MRI images from single-modality MRI data using Mamba. Our approach shows significant improvements in image quality, validated by strong Spearman correlation (R), high structural similarity index (SSIM), enhanced mutual information (MI), reduced normalized root mean square error (NRMSE), and lower symmetric mean absolute percent error (SMAPE). Results demonstrate that VMKANet and CAViT effectively enhance breast MRI images, promising safer and more affordable diagnostic capabilities.

KW - Attention

KW - Breast Cancer

KW - cGAN MRI Enhancement

KW - KAN

KW - Mamba

UR - https://www.scopus.com/pages/publications/105018800996

U2 - 10.1007/978-3-032-06624-4_16

DO - 10.1007/978-3-032-06624-4_16

M3 - Conference article published in proceeding or book

AN - SCOPUS:105018800996

SN - 9783032066237

VL - 16178

T3 - Lecture Notes in Computer Science

SP - 150

EP - 159

BT - Computational Mathematics Modeling in Cancer Analysis - 4th International Workshop, CMMCA 2025, Held in Conjunction with MICCAI 2025, Proceedings

A2 - Li, Chao

A2 - Qin, Wenjian

A2 - Wu, Jia

A2 - Zaki, Nazar

PB - Springer Science and Business Media Deutschland GmbH

T2 - 4th International Workshop on Computational Mathematics Modeling in Cancer Analysis, CMMCA 2025, Held in Conjunction with the 28th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2025

Y2 - 27 September 2025 through 27 September 2025

ER -

Pu Y, Zhu Y, Qu J, Wang X, Li W, Zhao M et al. Revealing New Possibilities for Breast MRI Enhancement: Mamba-Driven Cross-Attention GAN with VMKANet. In Li C, Qin W, Wu J, Zaki N, editors, Computational Mathematics Modeling in Cancer Analysis - 4th International Workshop, CMMCA 2025, Held in Conjunction with MICCAI 2025, Proceedings. Vol. 16178. Springer Science and Business Media Deutschland GmbH. 2025. p. 150-159. (Lecture Notes in Computer Science). doi: 10.1007/978-3-032-06624-4_16