DualSS: A Dual-Diffusion Synthetic Sampling Framework for Screening Fatty Liver Severity With Class-Imbalanced Tongue Images

Tao Chen , Jie Gao , Yong Xu , Weihong Qiu , Kunhong Liu , Yijie Wu , Weimin Ye

CAAI Transactions on Intelligence Technology ›› 2026, Vol. 11 ›› Issue (5) : 1511 -1528.

PDF (3605KB)
CAAI Transactions on Intelligence Technology ›› 2026, Vol. 11 ›› Issue (5) :1511 -1528. DOI: 10.1049/cit2.70174
ORIGINAL RESEARCH
research-article
DualSS: A Dual-Diffusion Synthetic Sampling Framework for Screening Fatty Liver Severity With Class-Imbalanced Tongue Images
Author information +
History +
PDF (3605KB)

Abstract

Fatty liver disease (FLD) is a prevalent chronic condition that can progress to clinically significant liver injury if left untreated. Given its high prevalence, FLD poses a substantial public health burden. Traditional Chinese medicine suggests that the appearance of tongue reflects hepatic and metabolic status, enabling artificial intelligence-based tongue diagnosis to provide rapid low-cost prescreening for FLD severity. In practice, the development of high-quality models is hindered by costly disease labelling, scarcity of positive cases, and severe class imbalance. In this work, we propose DualSS, a dual-diffusion synthetic sampling framework to mitigate these limitations. DualSS first trains a disease-agnostic dual-diffusion generator (DDG) on tongue-image data, leveraging abundant negative samples to learn unified tongue representations. Building on DDG, we design a latent space synthetic sampling pipeline that synthesises realistic positive-class images to construct a more balanced augmented training dataset. We then couple this pipeline with a noise-robust training strategy on the augmented dataset to further enhance the learning of downstream diagnostic models. Extensive experiments on FLD severity classification using tongue images and physiological indicators show that DualSS consistently outperforms existing rebalancing baselines, offering an effective and generalisable solution to alleviate positive sample scarcity in tongue image-based medical diagnosis.

Keywords

fatty liver disease / latent diffusion model / physics-informed neural networks / synthetic sampling / tongue diagnosis

Cite this article

Download citation ▾
Tao Chen, Jie Gao, Yong Xu, Weihong Qiu, Kunhong Liu, Yijie Wu, Weimin Ye. DualSS: A Dual-Diffusion Synthetic Sampling Framework for Screening Fatty Liver Severity With Class-Imbalanced Tongue Images. CAAI Transactions on Intelligence Technology, 2026, 11 (5) : 1511-1528 DOI:10.1049/cit2.70174

登录浏览全文

4963

注册一个新账户 忘记密码

References

[1]

M. H. Le, D. M. Le, T. C. Baez, et al., “ Global Incidence of Non-Alcoholic Fatty Liver Disease: A Systematic Review and Meta-Analysis of 63 Studies and 1,201,807 Persons,” Journal of Hepatology 79, no. 2 (2023): 287-295, https://doi.org/10.1016/j.jhep.2023.03.040.

[2]

A. Borrelli, P. Bonelli, F. M. Tuccillo, et al., “ Role of Gut Microbiota and Oxidative Stress in the Progression of Non-Alcoholic Fatty Liver Disease to Hepatocarcinoma: Current and Innovative Therapeutic Approaches,” Redox Biology 15 (2018): 467-479, https://doi.org/10.1016/j.redox.2018.01.009.

[3]

Y.-Y. Chen and M. M. Yeh, “ Non-Alcoholic Fatty Liver Disease: A Review With Clinical and Pathological Correlation,” Journal of the Formosan Medical Association 120, no. 1 Part 1 (2021): 68-77, https://doi.org/10.1016/j.jfma.2020.07.006.

[4]

S. K. Han, S. K. Baik, and M. Y. Kim, “ Non-Alcoholic Fatty Liver Disease: Definition and Subtypes,” supplement, Clinical and Molecular Hepatology 29, no. Suppl (2023): S5-S16, https://doi.org/10.3350/cmh.2022.0424.

[5]

H. Park, E. L. Yoon, M. Kim, et al., “ Cost-Effectiveness of Advanced Hepatic Fibrosis Screening in Individuals With Suspected MASLD Identified by Serologic Noninvasive Tests,” Scientific Reports 15, no. 1 (2025): 24186, https://doi.org/10.1038/s41598-025-08434-z.

[6]

Q. Liu, Y. Li, P. Yang, et al., “ A Survey of Artificial Intelligence in Tongue Image for Disease Diagnosis and Syndrome Differentiation,” Digital Health 9 (2023): 20552076231191044, https://doi.org/10.1177/20552076231191044.

[7]

S. Lee, B. Arefaine, N. Begum, et al., “ Oral-Gut Microbiome Interactions in Advanced Cirrhosis: Characterisation of Pathogenic Enterotypes and Salivatypes, Virulence Factors and Antimicrobial Resistance,” Journal of Hepatology 82, no. 4 (2025): 622-633, https://doi.org/10.1016/j.jhep.2024.09.046.

[8]

C. Lu, H. Zhu, D. Zhao, et al., “ Oral-Gut Microbiome Analysis in Patients With Metabolic-Associated Fatty Liver Disease Having Different Tongue Image Feature,” Frontiers in Cellular and Infection Microbiology 12 (2022): 787143, https://doi.org/10.3389/fcimb.2022.787143.

[9]

M.-C. Hu, K.-C. Lan, W.-C. Fang, et al., “ Automated Tongue Diagnosis on the Smartphone and Its Applications,” Computer Methods and Programs in Biomedicine 174 (2019): 51-64, https://doi.org/10.1016/j.cmpb.2017.12.029.

[10]

T. Jiang, X.-J. Guo, L.-P. Tu, et al., “ Application of Computer Tongue Image Analysis Technology in the Diagnosis of NAFLD,” Computers in Biology and Medicine 135 (2021): 104622, https://doi.org/10.1016/j.compbiomed.2021.104622.

[11]

R.-R. Wang, J.-L. Chen, S.-J. Duan, et al., “ Noninvasive Diagnostic Technique for Nonalcoholic Fatty Liver Disease Based on Features of Tongue Images,” Chinese Journal of Integrative Medicine 30, no. 3 (2024): 203-212, https://doi.org/10.1007/s11655-023-3616-1.

[12]

J. Gao, T. Chen, Y. Xu, et al., “ Accurate Fatty Liver Disease Diagnosis With a Multi-Source Feature Fusion Model on the Segmented Tongue Image Dataset,” Journal of Advanced Research 85 (2025): 507-520, (In Press), https://doi.org/10.1016/j.jare.2025.10.003.

[13]

T. Chen, J. Gao, Y. Xu, et al., “ Hard Sample Mining-Based Tongue Diagnosis for Fatty Liver Disease Severity Classification,” in Medical Image Computing and Computer Assisted Intervention - MICCAI 2025; Sep 23; Daejeon (Springer Nature Switzerland, 2025).

[14]

S. Man, Y. Deng, Y. Ma, et al., “ Prevalence of Liver Steatosis and Fibrosis in the General Population and Various High-Risk Populations: A Nationwide Study With 5.7 Million Adults in China,” Gastroenterology 165, no. 4 (2023): 1025-1040, https://doi.org/10.1053/j.gastro.2023.05.053.

[15]

S. Dai, X. Guo, S. Liu, et al., “ Application of Intelligent Tongue Image Analysis in Conjunction With Microbiomes in the Diagnosis of MAFLD,” Heliyon 10, no. 7 (2024): e29269, https://doi.org/10.1016/j.heliyon.2024.e29269.

[16]

T. Chen, Y. Chen, Z. Zhou, Y. Zhu, L. He, and J. Zhang, “ Deep Learning-Based Automated Tongue Analysis System for Assisted Chinese Medicine Diagnosis,” Frontiers in Physiology 16 (2025): 1559389, https://doi.org/10.3389/fphys.2025.1559389.

[17]

L. Yang, Q. Dong, D. Lin, and X. Lü, “ TongueNet: A Multi-Modal Fusion and Multi-Label Classification Model for Traditional Chinese Medicine Tongue Diagnosis,” Frontiers in Physiology 16 (2025): 1527751, https://doi.org/10.3389/fphys.2025.1527751.

[18]

X.-Z. Lu, S. Liu, X.-X. Lin, et al., “ An AI-Powered Tongue Image Model for Home-Based Monitoring of Liver Fibrosis,” npj Digital Medicine 9, no. 1 (2025): 67, https://doi.org/10.1038/s41746-025-02246-1.

[19]

M. Salmi, D. Atif, D. Oliva, A. Abraham, and S. Ventura, “ Handling Imbalanced Medical Datasets: Review of a Decade of Research,” Artificial Intelligence Review 57, no. 10 (2024): 273, https://doi.org/10.1007/s10462-024-10884-2.

[20]

W. Chen, K. Yang, Z. Yu, Y. Shi, and C. L. P. Chen, “ A Survey on Imbalanced Learning: Latest Research, Applications and Future Directions,” Artificial Intelligence Review 57, no. 6 (2024): 137, https://doi.org/10.1007/s10462-024-10759-6.

[21]

E. Pachetti and S. Colantonio, “ A Systematic Review of Few-Shot Learning in Medical Imaging,” Artificial Intelligence in Medicine 156 (2024): 102949, https://doi.org/10.1016/j.artmed.2024.102949.

[22]

T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, “ Focal Loss for Dense Object Detection,” in 2017 IEEE International Conference on Computer Vision (ICCV) (IEEE, 2017), 2999-3007.

[23]

Y. Cui, M. Jia, T.-Y. Lin, Y. Song, and S. Belongie, “ Class-Balanced Loss Based on Effective Number of Samples,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (IEEE, 2019), 9260-9269.

[24]

N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “ SMOTE: Synthetic Minority Over-Sampling Technique,” Journal of Artificial Intelligence Research 16, no. 1 (2002): 321-357, https://doi.org/10.1613/jair.953.

[25]

H. Han, W.-Y. Wang, and B.-H. Mao, “ Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning,” in Advances in Intelligent Computing (Springer Berlin Heidelberg, 2005), 878-887.

[26]

M. Han, H. Guo, J. Li, and W. Wang, “ Global-Local Information Based Oversampling for Multi-Class Imbalanced Data,” International Journal of Machine Learning and Cybernetics 14, no. 6 (2023): 2071-2086, https://doi.org/10.1007/s13042-022-01746-w.

[27]

D. Dablain, B. Krawczyk, and N. V. Chawla, “ DeepSMOTE: Fusing Deep Learning and SMOTE for Imbalanced Data,” IEEE Transactions on Neural Networks and Learning Systems 34, no. 9 (2023): 6390-6404, https://doi.org/10.1109/TNNLS.2021.3136503.

[28]

S. Li, Y. Lin, H. Chen, and K.-T. Cheng, “ Iterative Online Image Synthesis via Diffusion Model for Imbalanced Classification,” in Medical Image Computing and Computer Assisted Intervention-MICCAI 2024 (Springer Nature Switzerland, 2024), 371-381.

[29]

Z. Shen, M. Cao, S. Wang, L. Zhang, and Q. Wang, “ CellGAN: Conditional Cervical Cell Synthesis for Augmenting Cytopathological Image Classification,” in Medical Image Computing and Computer Assisted Intervention-MICCAI 2023 (Springer-Verlag, 2023), 487-496.

[30]

J. Ho, A. Jain, and P. Abbeel, “ Denoising Diffusion Probabilistic Models,” in Proceedings of the 34th International Conference on Neural Information Processing Systems (Curran Associates Inc., 2020), 574.

[31]

S. Rajaraman, Z. Liang, Z. Xue, and S. Antani, “ Addressing Class Imbalance With Latent Diffusion-Based Data Augmentation for Improving Disease Classification in Pediatric Chest X-Rays,” in 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) (IEEE, 2024), 5059-5066.

[32]

Y. Shi, A. Abulizi, H. Wang, et al., “ Diffusion Models for Medical Image Computing: A Survey,” Tsinghua Science and Technology 30, no. 1 (2025): 357-383, https://doi.org/10.26599/TST.2024.9010047.

[33]

Y. Qin, H. Zheng, J. Yao, M. Zhou, and Y. Zhang, “ Class-Balancing Diffusion Models,” in 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (IEEE, 2023), 18434-18443.

[34]

M. Raissi, P. Perdikaris, and G. E. Karniadakis, “ Physics-Informed Neural Networks: A Deep Learning Framework for Solving Forward and Inverse Problems Involving Nonlinear Partial Differential Equations,” Journal of Computational Physics 378 (2019): 686-707, https://doi.org/10.1016/j.jcp.2018.10.045.

[35]

S. Cuomo, V. S. Di Cola, F. Giampaolo, G. Rozza, M. Raissi, and F. Piccialli, “ Scientific Machine Learning Through Physics-Informed Neural Networks: Where We Are and What’s Next,” Journal of Scientific Computing 92, no. 3 (2022): 88, https://doi.org/10.1007/s10915-022-01939-z.

[36]

Z. Min, Z. M. C. Baum, S. U. Saeed, et al., “ Non-Rigid Medical Image Registration Using Physics-Informed Neural Networks,” in Information Processing in Medical Imaging; San Carlos de Bariloche, Argentina (Springer Nature Switzerland, 2023), 601-613.

[37]

F. Shone, N. Ravikumar, T. Lassila, et al., “ Deep Physics-Informed Super-Resolution of Cardiac 4D-Flow MRI,” in Information Processing in Medical Imaging (Springer Nature Switzerland, 2023), 511-522.

[38]

M. F. Fathi, I. Perez-Raya, A. Baghaie, et al., “ Super-Resolution and Denoising of 4D-Flow MRI Using Physics-Informed Deep Neural Nets,” Computer Methods and Programs in Biomedicine 197 (2020): 105729, https://doi.org/10.1016/j.cmpb.2020.105729.

[39]

M. Sarabian, H. Babaee, and K. Laksari, “ Physics-Informed Neural Networks for Brain Hemodynamic Predictions Using Medical Imaging,” IEEE Transactions on Medical Imaging 41, no. 9 (2022): 2285-2303, https://doi.org/10.1109/TMI.2022.3161653.

[40]

P. H. M. Voorter, W. H. Backes, O. J. Gurney-Champion, et al., “ Improving Microstructural Integrity, Interstitial Fluid, and Blood Microcirculation Images From Multi-b-Value Diffusion MRI Using Physics-Informed Neural Networks in Cerebrovascular Disease,” Magnetic Resonance in Medicine 90, no. 4 (2023): 1657-1671, https://doi.org/10.1002/mrm.29753.

[41]

S. Zakany and M. C. Milinkovitch, “ Simple Reaction-Diffusion Modeling Predicts Inconspicuous Neighborhood-Dependent Color Subclustering of Lizard Scales,” Physical Review X 13, no. 4 (2023): 041011, https://doi.org/10.1103/PhysRevX.13.041011.

[42]

A. Van Den Oord, O. Vinyals, and K. Kavukcuoglu, “ Neural Discrete Representation Learning,” in Advances in Neural Information Processing Systems (Curran Associates, Inc., 2017).

[43]

Z. Wang and Z. Liu, “ Posterior Collapse of a Linear Latent Variable Model,” in Advances in Neural Information Processing Systems (Curran Associates, Inc., 2022), 37537-37548.

[44]

R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “ High-Resolution Image Synthesis With Latent Diffusion Models,” in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (IEEE, 2022), 10684-10695.

[45]

L. Yuan, L. Yang, S. Zhang, et al., “ Development of a Tongue Image-Based Machine Learning Tool for the Diagnosis of Gastric Cancer: A Prospective Multicentre Clinical Cohort Study,” EClinicalMedicine 57 (2023): 101834, https://doi.org/10.1016/j.eclinm.2023.101834.

[46]

A. Vaswani, N. Shazeer, N. Parmar, et al., “ Attention Is all You Need,” in Proceedings of the 31st International Conference on Neural Information Processing Systems (Curran Associates Inc., 2017), 6000-6010.

[47]

D. L. Gandy and M. R. Nelson, “ Analyzing Pattern Formation in the Gray-Scott Model: An XPPAUT Tutorial,” SIAM Review 64, no. 3 (2022): 728-747, https://doi.org/10.1137/21M1402868.

[48]

M. Tahmid, M. S. Alam, N. Rao, and K. M. A. Ashrafi, “ Image-to-Image Translation With Conditional Adversarial Networks,” in 2023 IEEE 9th International Women in Engineering (WIE) Conference on Electrical and Computer Engineering (WIECON-ECE), (2023), 1-5s.

[49]

X. Ma, Y. Wang, M. E. Houle, et al., “ Dimensionality-Driven Learning With Noisy Labels,” in Proceedings of the 35th International Conference on Machine Learning; Proceedings of Machine Learning Research (PMLR, 2018), 3355-3364.

[50]

K. Cao, C. Wei, A. Gaidon, N. Arechiga, and T. Ma, “ Learning Imbalanced Datasets With Label-Distribution-Aware Margin Loss,” in Advances in Neural Information Processing Systems (Curran Associates, Inc., 2019).

[51]

M. Ren, W. Zeng, B. Yang, and R. Urtasun, “ Learning to Reweight Examples for Robust Deep Learning,” in Proceedings of the 35th International Conference on Machine Learning; Proceedings of Machine Learning Research (PMLR, 2018), 4334- 4343.

[52]

B. Kang, S. Xie, M. Rohrbach, et al., “ Decoupling Representation and Classifier for Long-Tailed Recognition,” arXiv Preprint (2019), https://doi.org/10.48550/arXiv.1910.09217.

[53]

X. Wang, X. Wang, Y. Lou, et al., “ Constructing Tongue Coating Recognition Model Using Deep Transfer Learning to Assist Syndrome Diagnosis and Its Potential in Noninvasive Ethnopharmacological Evaluation,” Journal of Ethnopharmacology 285 (2022): 114905, https://doi.org/10.1016/j.jep.2021.114905.

[54]

D. Qiu, X. Zhang, X. Wan, et al., “ A Novel Tongue Feature Extraction Method on Mobile Devices,” Biomedical Signal Processing and Control 80 (2023): 104271, https://doi.org/10.1016/j.bspc.2022.104271.

[55]

X. Sha, Z. Guan, Z. Chen, J. Han, and Y. Wang, “ SSC-Net: A Multi-Task Joint Learning Network for Tongue Image Segmentation and Multi-Label Classification,” Digital Health 11 (2025): 20552076251343696, https://doi.org/10.1177/20552076251343696.

[56]

P. Castellano-Escuder, D. K. Zachman, K. Han, and M. D. Hirschey, “ GAUDI: Interpretable Multi-Omics Integration With UMAP Embeddings and Density-Based Clustering,” Nature Communications 16, no. 1 (2025): 5771, https://doi.org/10.1038/s41467-025-60822-1.

PDF (3605KB)

0

Accesses

0

Citation

Detail

Sections
Recommended

/

〈 〉