Liver lesion classification in multi-modal magnetic resonance imaging via medical foundation segmentation models and radiomics integration_Journal of Biomedical Engineering

Authors：

LIU Yunpeng ¹ ,  WENCHENG Dangzhi ² , ZHU Jialu ¹ , WANG Yipeng ¹ , NIU Junmin ¹ , GAN Kaifeng ³

1. Information and Computing Science, Ningbo University of Technology, Ningbo, Zhejiang 315000, P. R. China;
2. School of Pharmacy, Qinghai Minzu University, Xining 810000, P. R. China;
3. Li Huili Hospital Affiliated to Ningbo University, Ningbo, Zhejiang 315000, P. R. China;

Corresponding?author：

WENCHENG Dangzhi, Email: 503534641@qq.com

Keywords：

Medical foundation segmentation models; Magnetic resonance imaging; Radiomics; Multi-modal fusion; Liver lesion classification

DOI：

10.7507/1001-5515.202505005

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

This study proposes a framework integrating medical foundation segmentation models and radiomics for liver lesion segmentation and classification in multi-modal magnetic resonance imaging (MRI). MedSAM2 and VISTA3D are complementarily fine-tuned and combined through spatial gated fusion to generate lesion masks. Deep semantic features extracted by M3D are integrated with handcrafted radiomics features through within-modality alignment, cross-modal interaction, and sparse selection. Negative sampling and an extreme gradient boosting (XGBoost) classifier are then used for lesion classification. On the validation set of the public LLD-MMRI2023 dataset, the framework achieved a Dice coefficient of 0.90, an F1 score of 0.837, and a Cohen’s Kappa of 0.792. The latter two metrics represented relative improvements of 8.27% and 7.57% over the best known public results. Ablation experiments demonstrated the contributions of joint fine-tuning, spatial priors, and multi-modal feature fusion. The proposed framework combines foundation segmentation models, deep semantic features, and radiomics to support liver lesion segmentation and classification in multi-modal MRI.

Citation： LIU Yunpeng, WENCHENG Dangzhi, ZHU Jialu, WANG Yipeng, NIU Junmin, GAN Kaifeng. Liver lesion classification in multi-modal magnetic resonance imaging via medical foundation segmentation models and radiomics integration. Journal of Biomedical Engineering, 2026, 43(5): 1018-1025. doi: 10.7507/1001-5515.202505005 Copy

1.	Gul S, Khan M S, Bibi A, et al. Deep learning techniques for liver and liver tumor segmentation: a review[J]. Comput Biol Med, 2022, 147: 105620.
2.	Saha Roy S, Roy S, Mukherjee P, et al. An automated liver tumour segmentation and classification model by deep learning based approaches[J]. Comput Methods Biomech Biomed Eng Imaging Vis, 2023, 11(3): 638-650.
3.	Saumiya S, Franklin S W. Unified automated deep learning framework for segmentation and classification of liver tumors[J]. J Supercomput, 2024, 80(2): 2347-2380.
4.	Malik M H, Ghous H, Rashid T, et al. Feature extraction-based liver tumor classification using machine learning and deep learning methods of computed tomography images[J]. Cogent Eng, 2024, 11(1): 2338994.
5.	Ma J, He Y, Li F, et al. Segment anything in medical images[J]. Nat Commun, 2024, 15(1): 654.
6.	Zhu J, Hamdi A, Qi Y, et al. Medical SAM 2: segment medical images as video via segment anything model 2[PP/OL]. arXiv (2024-12-04) [2025-03-05]. https://doi.org/10.48550/arXiv.2408.00874.
7.	Ma J, Li F, Kim S, et al. Efficient MedSAMs: segment anything in medical images on laptop[PP/OL]. arXiv (2024-12-20) [2025-03-05]. https://doi.org/10.48550/arXiv.2412.16085.
8.	Zhu Z, Ma X, Chen Y, et al. 3D-VISTA: pre-trained transformer for 3D vision and text alignment[C]//2023 IEEE/CVF International Conference on Computer Vision (ICCV). Paris: IEEE, 2023: 2911-2921.
9.	Zossou V B S, Gnangnon F H R, Biaou O, et al. Radiomics-based classification of tumor and healthy liver on computed tomography images[J]. Cancers, 2024, 16(6): 1158.
10.	Xie X Y, Chen R. Research progress of MRI-based radiomics in hepatocellular carcinoma[J]. Front Oncol, 2025, 15: 1420599.
11.	Bai F, Du Y, Huang T, et al. M3D: advancing 3D medical image analysis with multi-modal large language models[PP/OL]. arXiv (2024-03-31) [2025-03-05]. https://doi.org/10.48550/arXiv.2404.00578.
12.	Mao B, Zhang L, Ning P, et al. Preoperative prediction for pathological grade of hepatocellular carcinoma via machine learning-based radiomics[J]. Eur Radiol, 2020, 30(12): 6924-6932.
13.	Wu C, Du X, Zhang Y, et al. Five machine learning-based radiomics models for preoperative prediction of histological grade in hepatocellular carcinoma[J]. J Cancer Res Clin Oncol, 2023, 149(16): 15103-15112.
14.	Zhang H W, Huang D L, Wang Y R, et al. CT radiomics based on different machine learning models for classifying gross tumor volume and normal liver tissue in hepatocellular carcinoma[J]. Cancer Imaging, 2024, 24(1): 20.
15.	Li J, Zou L, Ma H, et al. Interpretable machine learning based on CT-derived extracellular volume fraction to predict pathological grading of hepatocellular carcinoma[J]. Abdom Radiol, 2024, 49(10): 3383-3396.
16.	Lou M, Ying H, Liu X, et al. SDR-Former: a siamese dual-resolution transformer for liver lesion classification using 3D multi-phase imaging[J]. Neural Netw, 2025, 185: 107228.
17.	LMMMEng. LLD-MMRI2023[EB/OL]. (2023) [2025-03-05]. https://github.com/LMMMEng/LLD-MMRI2023.
18.	Bowang-lab. MedSAM2[EB/OL]. (2025) [2025-03-05]. https://github.com/bowang-lab/MedSAM2.
19.	Wang Lab. LLD-MMRI-MedSAM2[EB/OL]. (2025) [2025-03-05]. https://huggingface.co/datasets/wanglab/LLD-MMRI-MedSAM2.
20.	Hatamizadeh A, Nath V, Tang Y, et al. Swin UNETR: Swin transformers for semantic segmentation of brain tumors in MRI images[C]//Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Cham: Springer, 2022: 272-284.
21.	Isensee F, Jaeger P F, Kohl S A A, et al. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation[J]. Nat Methods, 2021, 18(2): 203-211.
22.	Myronenko A. 3D MRI brain tumor segmentation using autoencoder regularization[C]//Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Cham: Springer, 2019: 311-320.
23.	LMMMEng. LLD-MMRI2023 validation leaderboard[EB/OL]. (2023) [2025-03-05]. https://github.com/LMMMEng/LLD-MMRI2023/blob/main/assets/validation_leaderboard.md.

1. Gul S, Khan M S, Bibi A, et al. Deep learning techniques for liver and liver tumor segmentation: a review[J]. Comput Biol Med, 2022, 147: 105620.
2. Saha Roy S, Roy S, Mukherjee P, et al. An automated liver tumour segmentation and classification model by deep learning based approaches[J]. Comput Methods Biomech Biomed Eng Imaging Vis, 2023, 11(3): 638-650.
3. Saumiya S, Franklin S W. Unified automated deep learning framework for segmentation and classification of liver tumors[J]. J Supercomput, 2024, 80(2): 2347-2380.
4. Malik M H, Ghous H, Rashid T, et al. Feature extraction-based liver tumor classification using machine learning and deep learning methods of computed tomography images[J]. Cogent Eng, 2024, 11(1): 2338994.
5. Ma J, He Y, Li F, et al. Segment anything in medical images[J]. Nat Commun, 2024, 15(1): 654.
6. Zhu J, Hamdi A, Qi Y, et al. Medical SAM 2: segment medical images as video via segment anything model 2[PP/OL]. arXiv (2024-12-04) [2025-03-05]. https://doi.org/10.48550/arXiv.2408.00874.
7. Ma J, Li F, Kim S, et al. Efficient MedSAMs: segment anything in medical images on laptop[PP/OL]. arXiv (2024-12-20) [2025-03-05]. https://doi.org/10.48550/arXiv.2412.16085.
8. Zhu Z, Ma X, Chen Y, et al. 3D-VISTA: pre-trained transformer for 3D vision and text alignment[C]//2023 IEEE/CVF International Conference on Computer Vision (ICCV). Paris: IEEE, 2023: 2911-2921.
9. Zossou V B S, Gnangnon F H R, Biaou O, et al. Radiomics-based classification of tumor and healthy liver on computed tomography images[J]. Cancers, 2024, 16(6): 1158.
10. Xie X Y, Chen R. Research progress of MRI-based radiomics in hepatocellular carcinoma[J]. Front Oncol, 2025, 15: 1420599.
11. Bai F, Du Y, Huang T, et al. M3D: advancing 3D medical image analysis with multi-modal large language models[PP/OL]. arXiv (2024-03-31) [2025-03-05]. https://doi.org/10.48550/arXiv.2404.00578.
12. Mao B, Zhang L, Ning P, et al. Preoperative prediction for pathological grade of hepatocellular carcinoma via machine learning-based radiomics[J]. Eur Radiol, 2020, 30(12): 6924-6932.
13. Wu C, Du X, Zhang Y, et al. Five machine learning-based radiomics models for preoperative prediction of histological grade in hepatocellular carcinoma[J]. J Cancer Res Clin Oncol, 2023, 149(16): 15103-15112.
14. Zhang H W, Huang D L, Wang Y R, et al. CT radiomics based on different machine learning models for classifying gross tumor volume and normal liver tissue in hepatocellular carcinoma[J]. Cancer Imaging, 2024, 24(1): 20.
15. Li J, Zou L, Ma H, et al. Interpretable machine learning based on CT-derived extracellular volume fraction to predict pathological grading of hepatocellular carcinoma[J]. Abdom Radiol, 2024, 49(10): 3383-3396.
16. Lou M, Ying H, Liu X, et al. SDR-Former: a siamese dual-resolution transformer for liver lesion classification using 3D multi-phase imaging[J]. Neural Netw, 2025, 185: 107228.
17. LMMMEng. LLD-MMRI2023[EB/OL]. (2023) [2025-03-05]. https://github.com/LMMMEng/LLD-MMRI2023.
18. Bowang-lab. MedSAM2[EB/OL]. (2025) [2025-03-05]. https://github.com/bowang-lab/MedSAM2.
19. Wang Lab. LLD-MMRI-MedSAM2[EB/OL]. (2025) [2025-03-05]. https://huggingface.co/datasets/wanglab/LLD-MMRI-MedSAM2.
20. Hatamizadeh A, Nath V, Tang Y, et al. Swin UNETR: Swin transformers for semantic segmentation of brain tumors in MRI images[C]//Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Cham: Springer, 2022: 272-284.
21. Isensee F, Jaeger P F, Kohl S A A, et al. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation[J]. Nat Methods, 2021, 18(2): 203-211.
22. Myronenko A. 3D MRI brain tumor segmentation using autoencoder regularization[C]//Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Cham: Springer, 2019: 311-320.
23. LMMMEng. LLD-MMRI2023 validation leaderboard[EB/OL]. (2023) [2025-03-05]. https://github.com/LMMMEng/LLD-MMRI2023/blob/main/assets/validation_leaderboard.md.

Journal of Biomedical Engineering

Liver lesion classification in multi-modal magnetic resonance imaging via medical foundation segmentation models and radiomics integration

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content