AccScience Publishing / EJMO / Online First / DOI: 10.36922/EJMO025260276
Submit to EJMO
Cite this article
6
Download
75
Views
Journal Browser
Volume | Year
Issue
Search
Forthcoming Issue
Current Issue
View All
News and Announcements
View All
ORIGINAL RESEARCH ARTICLE

A comparative analysis of U-Net-based architectures for robust segmentation of bladder cancer lesions in magnetic resonance imaging

Ishak Pacal1,2* ,  Yigitcan Cakmak1,2
Show Less
1 Department of Computer Engineering, Faculty of Engineering, Igdir University, Igdir, Türkiye
2 Department of Electronics and Information Technologies, Faculty of Architecture and Engineering, Nakhchivan State University, Nakhchivan, Azerbaijan
EJMO, 025260276 https://doi.org/10.36922/EJMO025260276
Received: 26 June 2025 | Revised: 14 September 2025 | Accepted: 22 September 2025 | Published online: 27 October 2025
(This article belongs to the Special Issue New Developments in Bladder Cancer Treatment and Management)
© 2025 by the Author(s). This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution -Noncommercial 4.0 International License (CC-by the license) ( https://creativecommons.org/licenses/by-nc/4.0/ )
Download PDF
XML
Cite
Abstract

Introduction: Bladder cancer (BCa) represents a significant uro-oncological challenge due to its aggressive nature and high recurrence rates. Although magnetic resonance imaging (MRI) is a cornerstone modality in BCa management, the manual segmentation of lesions is time-consuming and suffers from low reproducibility due to inter- and intra-observer variability, morphological heterogeneity, and MRI artifacts.

Objective: This study aims to address these limitations by conducting a rigorous comparative evaluation of four distinct U-Net-based deep learning architectures.

Methods: The models were evaluated using the publicly available, multi-center FedBCa dataset, comprising 275 T2-weighted MRI scans from 228 patients. Using a standardized training protocol, performance was rigorously assessed with a suite of quantitative metrics, including the Dice coefficient, intersection over union (IoU), and Hausdorff distance, supplemented by qualitative visual comparison.

Results: Cross-scale mixer U-Net (CMUNet) achieved the best overall performance, yielding the highest Dice coefficient (0.7937), IoU (0.7033), and boundary delineation accuracy (Hausdorff distance: 8.4550 mm. Architectural trade-offs were evident: CMUNeXt was the most computationally efficient and offered the highest lesion sensitivity (0.9656), whereas Attention U-Net recorded the highest precision (0.8380).

Conclusion: CMUNet provides the most balanced and accurate performance for BCa segmentation. However, the optimal architecture choice is application-dependent; high-sensitivity models such as CMUNeXt are ideal for screening, while high-precision models like Attention U-Net are better suited for treatment planning. Deep learning models serve as powerful assistive tools to improve efficiency and objectivity in clinical workflows, though expert oversight remains essential. The top model’s accuracy approached, but did not surpass, the inter-rater reliability of human experts (Dice: 0.870).

Keywords
Bladder cancer
Deep learning
U-Net
Image segmentation
Magnetic resonance imaging
Clinical decision support systems
Funding
None.
Conflict of interest
The authors declare that they have no competing interests.
References
  1. Bashkami A, Nasayreh A, Makhadmeh SN, et al. A review of artificial intelligence methods in bladder cancer: Segmentation, classification, and detection. Artif Intell Rev. 2024;57:339. doi: 10.1007/s10462-024-10953-6

 

  1. Li L, Jiang L, Yang K, Luo B, Wang X. A novel artificial intelligence segmentation model for early diagnosis of bladder tumors. Abdom Radiol. 2024;50(7):3092-3099. doi: 10.1007/S00261-024-04715-9

 

  1. Wang S, Wei Z, Shu H, et al. Early diagnosis and prognostic potential of RAC3 in bladder tumor. Int Urol Nephrol. 2024;56(2):475-482. doi: 10.1007/S11255-023-03781-0

 

  1. Gercek O, Ulusoy K, Yazar VM, Topal K. Effects of delayed diagnosis on tumor size, stage and grade in bladder cancer. Int Urol Nephrol. 2024;56(3):935-940. doi: 10.1007/S11255-023-03829-1

 

  1. Nasrallah AA, Williams SB. Epidemiology of bladder cancer. In: Biology of Bladder Cancer. Berlin: Springer; 2024. p. 3-22. doi: 10.1007/978-3-031-68505-7_1

 

  1. Wang Y, Zhou H, Tang Z, et al. Smoking and bladder cancer: Insights into pathogenesis and public health implications from a bibliometric analysis (1999-2023). Subst Abuse Treat Prev Policy. 2025;20(1):25. doi: 10.1186/S13011-025-00658-6

 

  1. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin. 2024;74:12-49. doi: 10.3322/caac.21820

 

  1. He C, Xu H, Yuan E, et al. The accuracy and quality of image-based artificial intelligence for muscle-invasive bladder cancer prediction. Insights Imaging. 2024;15(1):185. doi: 10.1186/S13244-024-01780-Y

 

  1. Gnyawali D, Sigdel PR, Rai BDK, Chapagain S, Gyawali PR. Storz professional image enhancement system (SPECTRA A) enhancing detection of carcinoma urinary bladder by white light cystoscopy. Afr J Urol. 2024;30(1):27. doi: 10.1186/S12301-024-00429-Y

 

  1. Badawy MA, Zaky SM, Korkor MS, Fattah SA. Role of intravoxel incoherent motion MRI in diagnosis and assessment of invasiveness of urothelial bladder carcinoma. Egyptian Journal of Radiology and Nuclear Medicine. 2024;55(1):40. doi: 10.1186/S43055-024-01212-3

 

  1. Zhang R, Jia S, Zhai L, Wu F, Zhang S, Li F. Predicting preoperative muscle invasion status for bladder cancer using computed tomography-based radiomics nomogram. BMC Med Imaging. 2024;24(1):98. doi: 10.1186/S12880-024-01276-7

 

  1. Arita Y, Kwee TC, Akin O, et al. Multiparametric MRI and artificial intelligence in predicting and monitoring treatment response in bladder cancer. Insights Imaging. 2025;16(1):7. doi: 10.1186/S13244-024-01884-5

 

  1. Donswijk ML, Ettema RH, Meijer D, et al. The accuracy and intra- and interobserver variability of PSMA PET/CT for the local staging of primary prostate cancer. Eur J Nucl Med Mol Imaging. 2024;51(6):1741-1752. doi: 10.1007/S00259-024-06594-0

 

  1. Ozdemir B, Sermet F, Pacal I. Attention-enhanced ConvNeXt for accurate, efficient, and interpretable crack detection. Expert Syst Appl. 2026;296:129165. doi: 10.1016/J.ESWA.2025.129165

 

  1. Pacal I, Kilicarslan S, Ozdemir B, Deveci M, Kadry S. Efficient and autonomous detection of olive leaf diseases using AI-enhanced MetaFormer. Artif Intell Rev. 2025;58:303. doi: 10.1007/S10462-025-11131-Y

 

  1. Cakmak Y. Machine learning approaches for enhanced diagnosis of hematological disorders. Comput Syst Artif Intell. 2025;1(1):8-14. doi: 10.69882/ADBA.CSAI.2025072

 

  1. Pacal I. Investigating deep learning approaches for cervical cancer diagnosis: A focus on modern image-based models. Eur J Gynaecol Oncol. 2025;46(1):125-141. doi: 10.22514/ejgo.2025.012

 

  1. Cakmak Y, Maman A. Deep learning for early diagnosis of lung cancer. Comput Syst Artif Intell. 2025;1(1):20-25. doi: 10.69882/ADBA.CSAI.2025074

 

  1. Cakmak Y, Pacal N. Deep learning for automated breast cancer detection in ultrasound: A comparative study of four CNN architectures. Artif Intell Appl Sci. 2025;1(1):13-19. doi: 10.69882/ADBA.AI.2025073

 

  1. Cakmak Y, Zeynalov J. A comparative analysis of convolutional neural network architectures for breast cancer classification from mammograms. Artif Intell Appl Sci. 2025;1(1):28-34. doi: 10.69882/ADBA.AI.2025075

 

  1. Pacal I, Attallah O. Hybrid deep learning model for automated colorectal cancer detection using local and global feature extraction. Knowl Based Syst. 2025;319:113625. doi: 10.1016/J.KNOSYS.2025.113625

 

  1. Pacal I, Attallah O. InceptionNeXt-Transformer: A novel multi-scale deep feature learning architecture for multimodal breast cancer diagnosis. Biomed Signal Process Control. 2025;110:108116. doi: 10.1016/J.BSPC.2025.108116

 

  1. Pacal I, Akhan O, Deveci RT, Deveci M. NeXtBrain: Combining local and global feature learning for brain tumor classification. Brain Res. 2025;1863:149762. doi: 10.1016/J.BRAINRES.2025.149762

 

  1. Aruk I, Pacal I, Toprak AN. A novel hybrid ConvNeXt-based approach for enhanced skin lesion classification. Expert Syst Appl. 2025;283:127721. doi: 10.1016/J.ESWA.2025.127721

 

  1. Ince S, Kunduracioglu I, Algarni A, Bayram B, Pacal I. Deep learning for cerebral vascular occlusion segmentation: A novel ConvNeXtV2 and GRN-integrated U-Net framework for diffusion-weighted imaging. Neuroscience. 2025;574:42-53. doi: 10.1016/J.NEUROSCIENCE.2025.04.010

 

  1. Dahbi H, Lazaar M, Hssayni EH. Deep learning based approach for medical image segmentation: A systematic literature review. In: Lecture Notes in Networks and Systems. Berlin: Springer; 2024. p. 1-16. doi: 10.1007/978-3-031-74491-4_1

 

  1. Kumar RR, Priyadarshi R. Denoising and segmentation in medical image analysis: A comprehensive review on machine learning and deep learning approaches. Multimed Tools Appl. 2024;84(12):10817-10875. doi: 10.1007/S11042-024-19313-6

 

  1. Adeel Abid M, Munir K. A systematic review on deep learning implementation in brain tumor segmentation, classification and prediction. In: Multimedia Tools and Applications. 2025. Berlin: Springer; 2025. p. 1-40. doi: 10.1007/S11042-025-20706-4

 

  1. Sherifa CA, Kanisha B. A Review on Medical Image Segmentation using Deep Learning. Berlin: Springer; 2024. p. 65-77. doi: 10.1007/978-981-97-5791-6_6

 

  1. Jain S, Pal A, Andrew J, Maheshwari S. Adaptive treatment planning via multi-class segmentation of GI tract tumours. In: Lecture Notes in Networks and Systems. Vol. 1122. Berlin: Springer; 2025. p. 477-492. doi: 10.1007/978-981-97-7426-5_36

 

  1. Rasool N, Bhat JI. A critical review on segmentation of glioma brain tumor and prediction of overall survival. Arch Comput Methods Eng. 2024;32(3):1525-1569. doi: 10.1007/S11831-024-10188-2

 

  1. Seeram E, Kanade V. Image processing and analysis. In: Artificial Intelligence in Medical Imaging Technology. Berlin: Springer; 2024. p. 83-103. doi: 10.1007/978-3-031-64049-0_5

 

  1. Do Nascimento Silva JH, Idrobo Pizo GA. Unsupervised bladder segmentation in cone beam-CT imaging via U-net model. In: 1st IFMBE Latin American Conference on Digital Health. Berlin: Springer; 2025. p. 28-46. doi: 10.1007/978-3-031-88064-3_3

 

  1. Das S, Sanki P, Mondal S, Bhattacharya S, Ghosh S. Bladder segmentation in MRI for high dose rate brachytherapy using deep network. In: Machine Learning for Social Transformation. Berlin: Springer; 2024. p. 55-67. doi: 10.1007/978-981-97-7532-3_5

 

  1. Zhao X, Lai L, Li Y, et al. A lightweight bladder tumor segmentation method based on attention mechanism. Med Biol Eng Comput. 2024;62(5):1519-1534. doi: 10.1007/S11517-024-03018-X

 

  1. Ye Z, Li Y, Sun Y, He C, He G, Ji Z. Leveraging deep learning in real-time intelligent bladder tumor detection during cystoscopy: A diagnostic study. Ann Surg Oncol. 2025;32(5):3220-3226.doi: 10.1245/S10434-025-17015-3

 

  1. Takagi H, Takeda K, Kadoya N, et al. Development of deep learning-based novel auto-segmentation for the prostatic urethra on planning CT images for prostate cancer radiotherapy. Radiol Phys Technol. 2024;17(4):819-826. doi: 10.1007/S12194-024-00832-8

 

  1. Mei H, Yang R, Huang J, et al. Artificial intelligence-assisted segmentation of prostate tumors and neurovascular bundles: Applications in precision surgery for prostate cancer. Ann Surg Oncol. 2025;18:1-9. doi: 10.1245/S10434-025-17659-1

 

  1. Wang Y, Wang H, Li N, et al. Unsupervised learning-based quantitative analysis of CT intratumoral subregions predicts risk stratification of bladder cancer patients. BMC Med. 2025;23(1):328. doi: 10.1186/S12916-025-04163-2

 

  1. Sun R, Zhang M, Yang L, et al. Preoperative CT-based deep learning radiomics model to predict lymph node metastasis and patient prognosis in bladder cancer: A two-center study. Insights Imaging. 2024;15(1):21. doi: 10.1186/s13244-023-01569-5

 

  1. Wei Z, Liu H, Xv Y, et al. Development and validation of a CT-based deep learning radiomics nomogram to predict muscle invasion in bladder cancer. Heliyon. 2024;10(2):e24878. doi: 10.1016/j.heliyon.2024.e24878

 

  1. Kurata Y, Nishio M, Moribata Y, et al. Development of deep learning model for diagnosing muscle-invasive bladder cancer on MRI with vision transformer. Heliyon. 2024;10(16):e36144. doi: 10.1016/j.heliyon.2024.e36144

 

  1. Han X, Guan J, Guo L, et al. A CT-based interpretable deep learning signature for predicting PD-L1 expression in bladder cancer: A two-center study. Cancer Imaging. 2025;25(1):27. doi: 10.1186/s40644-025-00849-1

 

  1. Gumus KZ, Nicolas J, Gopireddy DR, Dolz J, Jazayeri SB, Bandyk M. Deep learning algorithms for bladder cancer segmentation on multi-parametric MRI. Cancers (Basel). 2024;16(13):2348. doi: 10.3390/cancers16132348

 

  1. Jiang F, Hong G, Zeng H, et al. Deep learning-based model for prediction of early recurrence and therapy response on whole slide images in non-muscle-invasive bladder cancer: A retrospective, multicentre study. EClinicalMedicine. 2025;81:103125. doi: 10.1016/j.eclinm.2025.103125

 

  1. Wang GY, Zhu JF, Wang QC, et al. Prediction of non-muscle invasive bladder cancer recurrence using deep learning of pathology image. Sci Rep. 2024;14(1):18931. doi: 10.1038/s41598-024-66870-9

 

  1. Jiao P, Zheng Q, Yang R, et al. Prediction of HER2 status based on deep learning in H&E-stained histopathology images of bladder cancer. Biomedicines. 2024;12(7):1583. doi: 10.3390/biomedicines12071583

 

  1. Zheng Z, Dai F, Liu J, et al. Pathology-based deep learning features for predicting basal and luminal subtypes in bladder cancer. BMC Cancer. 2025;25(1):310. doi: 10.1186/s12885-025-13688-x

 

  1. Weng W, Zhu X. INet: Convolutional networks for biomedical image segmentation. IEEE Access. 2021;9:16591-16603. doi: 10.1109/ACCESS.2021.3053408

 

  1. Tang F, Wang L, Ning C, Xian M, Ding J. CMU-Net: A Strong Convmixer-Based Medical Ultrasound Image Segmentation Network. United States: Cornell University; 2022. doi: 10.1109/ISBI53787.2023.10230609

 

  1. Oktay O, Schlemper J, Le Folgoc L, et al. Attention U-Net: Learning Where to Look for the Pancreas. United States: Cornell University; 2018.

 

  1. Tang F, Ding J, Wang L, Ning C, Zhou SK. CMUNeXt: An Efficient Medical Image Segmentation Network Based on Large Kernel and Skip Fusion. Available from: https:// www.github.com/fenghetan9/cmunext [Last accessed on 2025 Jun 23].

 

  1. Cao K, Zou Y, Zhang C, et al. A multicenter bladder cancer MRI dataset and baseline evaluation of federated learning in clinical application. Sci Data. 2024;11(1):1147. doi: 10.1038/s41597-024-03971-0

 

  1. Cao K, Zou Y, Zhang C, et al. A Multi-Center MRI Dataset for Bladder Cancer and Baseline Evaluations of Federated Learning in its Clinical Application. Switzerland: Zenodo; 2024. doi: 10.5281/ZENODO.10409145

 

  1. Matuszczak M, Kiljańczyk A, Salagierski M. A liquid biopsy in bladder cancer-the current landscape in urinary biomarkers. Int J Mol Sci. 2022;23(15):8597. doi: 10.3390/IJMS23158597
Next article in this issue
Share
Back to top
Eurasian Journal of Medicine and Oncology, Electronic ISSN: 2587-196X Print ISSN: 2587-2400, Published by AccScience Publishing
We use cookies on our website to ensure you get the best experience.
Read more about our cookies here
Accept