Deep Learning-Based Segmentation of Hepatic Structures: A Systematic Review of Model Architectures and Clinical Applications

Autores

DOI:

https://doi.org/10.29384/rbfm.2026.v20.19849001882

Palavras-chave:

Liver cancer, Hepatic segmentation, Deep learning, Artificial intelligence in healthcare

Resumo

O câncer de fígado é um importante problema de saúde pública e permanece entre as neoplasias mais incidentes no mundo, com aproximadamente 866 mil novos casos em 2022, sendo o carcinoma hepatocelular responsável pela maior parte dos diagnósticos. A caracterização anatômica do fígado e de suas lesões depende, em grande parte, de imagens de tomografia computadorizada (TC) e ressonância magnética (RM), cuja segmentação ainda é majoritariamente manual ou semiautomática, o que exige tempo e pode gerar variações entre diferentes observadores ou até pelo mesmo profissional em momentos diferentes. Essa limitação motiva a adoção de métodos baseados em Deep Learning (DL), especialmente para aprimorar a precisão e a reprodutibilidade das segmentações hepáticas. Esta revisão analisou 16 estudos selecionados a partir de palavras-chave e critérios de inclusão predefinidos, com o objetivo de identificar tendências metodológicas, arquiteturas predominantes e aplicações clínicas reportadas. Observou-se forte predominância de modelos baseados em U-Net e variações (9/16), com crescente utilização de abordagens 3D (11/16) e foco em segmentações vasculares (6/16) ou fígado associado a vasos (6/16). As estratégias metodológicas dividiram-se igualmente entre modelos combinados (8/16) e abordagens de passo único (8/16). Apesar do desempenho técnico elevado relatado por vários autores, os estudos apresentaram heterogeneidade quanto aos protocolos de aquisição, métricas utilizadas e tamanho das bases de dados, dificultando comparações diretas e a extrapolação dos resultados para a prática clínica. Conclui-se que, embora as DLs representem um avanço significativo na segmentação hepática, a padronização metodológica e a validação externa dos modelos são etapas necessárias para sua consolidação no ambiente clínico.

Downloads

Não há dados estatísticos.

Referências

Khan N, Ahmed I, Kiran M, Adnan A. Overview of technical elements of liver segmentation. International Journal of Advanced Computer Science and Applications. 2016;7(12).

2. Forner A, Reig M, Bruix J. Hepatocellular carcinoma. The Lancet [Internet]. 2018;391(10127):1301–14. Available from: https://www.sciencedirect.com/science/article/pii/S0140673618300102

3. World Cancer Research Fund. Liver cancer. 2025 [cited 2025 Oct 15]. World Cancer Research Fund. Available from: https://www.wcrf.org/preventing-cancer/cancer-types/liver-cancer/

4. Zhou J, Sun H, Wang Z, Cong W, Zeng M, Zhou W, et al. Guidelines for the diagnosis and treatment of primary liver cancer (2022 edition). Liver Cancer. 2023;12(5):405–44.

5. Hu P, Wu F, Peng J, Liang P, Kong D. Automatic 3D liver segmentation based on deep learning and globally optimized surface evolution. Phys Med Biol. 2016;61(24):8676.

6. Pham DD, Dovletov G, Pauli J. Liver segmentation in CT with MRI data: zero-shot domain adaptation by contour extraction and shape priors. In: 2020 IEEE 17th international symposium on biomedical imaging (ISBI). IEEE; 2020. p. 1538–42.

7. Fabijańska A, Vacavant A, Lebre MA, Pavan ALM, de Pina DR, Abergel A, et al. U-CatcHCC: An accurate HCC detector in hepatic DCE-MRI sequences based on an U-Net framework. In: International Conference on Computer Vision and Graphics. Springer; 2018. p. 319–28.

8. Kashala Kabe G, Song Y, Liu Z. FireNet-MLstm for classifying liver lesions by using deep features in CT images. Multimed Tools Appl. 2022 Jan 1;81(2):1607–23.

9. Zheng R, Wang Q, Lv S, Li C, Wang C, Chen W, et al. Automatic liver tumor segmentation on dynamic contrast enhanced MRI using 4D information: deep learning model based on 3D convolution and convolutional LSTM. IEEE Trans Med Imaging. 2022;41(10):2965–76.

10. Ahmad M, Qadri SF, Qadri S, Saeed IA, Zareen SS, Iqbal Z, et al. A lightweight convolutional neural network model for liver segmentation in medical diagnosis. Comput Intell Neurosci. 2022;2022(1):7954333.

11. Liu X, Song L, Liu S, Zhang Y. A review of deep-learning-based medical image segmentation methods. Sustainability (Switzerland). 2021 Feb 1;13(3):1–29.

12. Hennedige T, Venkatesh SK. Imaging of hepatocellular carcinoma: diagnosis, staging and treatment monitoring. Cancer Imaging. 2013;12(3):530.

13. Gil AC. Como elaborar projetos de pesquisa. Vol. 4. Atlas São Paulo; 2002.

14. Ibragimov B, Toesca D, Chang D, Koong A, Xing L. Combining deep learning with anatomical analysis for segmentation of the portal vein for liver SBRT planning. Phys Med Biol. 2017;62(23):8943.

15. Roth HR, Oda H, Zhou X, Shimizu N, Yang Y, Hayashi Y, et al. An application of cascaded 3D fully convolutional networks for medical image segmentation. Computerized Medical Imaging and Graphics. 2018; 66:90–9.

16. Ahn Y, Yoon JS, Lee SS, Suk HI, Son JH, Sung YS, et al. Deep learning algorithm for automated segmentation and volume measurement of the liver and spleen using portal venous phase computed tomography images. Korean J Radiol. 2020;21(8):987.

17. Zbinden L, Catucci D, Suter Y, Berzigotti A, Ebner L, Christe A, et al. Convolutional neural network for automated segmentation of the liver and its vessels on non-contrast T1 vibe Dixon acquisitions. Sci Rep. 2022;12(1):22059.

18. Zbinden L, Catucci D, Suter Y, Hulbert L, Berzigotti A, Brönnimann M, et al. Automated liver segmental volume ratio quantification on non-contrast T1–Vibe Dixon liver MRI using deep learning. Eur J Radiol. 2023; 167:111047.

19. Alirr OI, Rahni AAA. Hepatic vessels segmentation using deep learning and preprocessing enhancement. J Appl Clin Med Phys. 2023;24(5): e13966.

20. Lee IC, Tsai YP, Lin YC, Chen TC, Yen CH, Chiu NC, et al. A hierarchical fusion strategy of deep learning networks for detection and segmentation of hepatocellular carcinoma from computed tomography images. Cancer Imaging. 2024;24(1):43.

21. Chen W, Zhao L, Bian R, Li Q, Zhao X, Zhang M. Compensation of small data with large filters for accurate liver vessel segmentation from contrast-enhanced CT images. BMC Med Imaging. 2024;24(1):129.

22. Gupta AC, Cazoulat G, Al Taie M, Yedururi S, Rigaud B, Castelo A, et al. Fully automated deep learning-based auto-contouring of liver segments and spleen on contrast-enhanced CT images. Sci Rep. 2024;14(1):4678.

23. Tanahashi Y, Kubota K, Nomura T, Ikeda T, Kutsuna M, Funayama S, et al. Improved vascular depiction and image quality through deep learning reconstruction of CT hepatic arteriography during transcatheter arterial chemoembolization. Jpn J Radiol. 2024;42(11):1243–54.

24. Li S, Li X, Zhou F, Zhang Y, Bie Z, Cheng L, et al. Automated segmentation of liver and hepatic vessels on portal venous phase computed tomography images using a deep learning algorithm. J Appl Clin Med Phys. 2024;25(8): e14397.

25. Kock F, Thielke F, Abolmaali N, Meine H, Schenk A. Suitability of DNN-based vessel segmentation for SIRT planning. Int J Comput Assist Radiol Surg. 2024;19(2):233–40.

26. Raab F, Strotzer Q, Stroszczynski C, Fellner C, Einspieler I, Haimerl M, et al. Automatic segmentation of liver structures in multi-phase MRI using variants of nnU-Net and Swin UNETR. Sci Rep. 2025;15(1):25740.

27. Cavicchioli M, Moglia A, Garret G, Puglia M, Vacavant A, Pugliese G, et al. D2-RD-UNet: A dual-stage dual-class framework with connectivity correction for hepatic vessels segmentation. Comput Biol Med. 2025; 195:110530.

28. Zheng T, Zhu Y, Jiang H, Yang C, Ye Y, Bashir MR, et al. MRI‐Based Topology Deep Learning Model for Noninvasive Prediction of Microvascular Invasion and Assisting Prognostic Stratification in HCC. Liver International. 2025;45(3): e16205.

29. Herold A, Sobotka D, Beer L, Bastati N, Poetter-Lang S, Weber M, et al. MRI-derived quantification of hepatic vessel-to-volume ratios in chronic liver disease using a deep learning approach. Eur Radiol Exp. 2025;9(1):75.

Downloads

Publicado

2026-05-07

Como Citar

Souza Dias Pinheiro, M., Massato Hanashiro Sakaguchi, R., Guilherme Soares Marques, T., Alexia Camargo Guassu, R., & Pina, D. R. de. (2026). Deep Learning-Based Segmentation of Hepatic Structures: A Systematic Review of Model Architectures and Clinical Applications. Revista Brasileira De Física Médica, 20, 882. https://doi.org/10.29384/rbfm.2026.v20.19849001882

Edição

Seção

Artigo de Revisão

Artigos Semelhantes

1 2 3 4 5 6 7 8 9 10 > >> 

Você também pode iniciar uma pesquisa avançada por similaridade para este artigo.

Artigos mais lidos pelo mesmo(s) autor(es)