U-NET BASED POLYP SEGMENTATION ON KVASIR-SEG DATASET: PERFORMANCE EVALUATION AND COMPARISON WITH STATE-OF-THE-ART METHODS
DOI:
https://doi.org/10.5281/zenodo.17874308Keywords:
U-Net, Polyp Segmentation, Kvasir-SEG, Deep Learning, Medical Image Segmentation, Convolutional Neural Networks.Abstract
Medical image segmentation is a fundamental task in computer-assisted diagnosis, enabling accurate
delineation of anatomical structures and pathological regions. In gastrointestinal endoscopy, precise segmentation of
polyps is crucial for early detection and treatment planning. This study presents a U-Net-based approach for polyp
segmentation using the publicly available Kvasir-SEG dataset. The proposed model was trained for 100 epochs with
custom metrics including Dice coefficient, Intersection over Union (IoU), precision, and recall. Experimental results
demonstrate that the model achieved an average Dice score of 0.9449, IoU of 0.9084, precision of 0.9404, and recall of
0.9584, outperforming several recent state-of-the-art methods. Comparative analysis with existing approaches confirms
the effectiveness of the vanilla U-Net architecture when combined with careful preprocessing and hyperparameter tuning.
The findings highlight U-Net's continued relevance in medical image segmentation tasks and suggest directions for future
work, including integration with attention mechanisms and transformer-based architectures.
References
S. Anantharajan, S. Gunasekaran, T. Subramanian, and V. R, “MRI brain tumor detection using deep learning
and machine learning approaches,” Meas. Sensors, vol. 31, no. January 2023, p. 101026, 2024, doi: 10.1016/j.
measen.2024.101026.
M. Arabboev, “BRAIN TUMOR CLASSIFICATION USING TRANSFER LEARNING WITH MOBILENETV2,” Top.
Issues Tech. Sci., vol. 3, no. 5, pp. 51–63, 2025, doi: https://doi.org/10.47390/ts-v3i5y2025N8.
P. Kaur and P. Mahajan, “Detection of brain tumors using a transfer learning-based optimized ResNet152 model in
MR images,” Comput. Biol. Med., vol. 188, no. February, p. 109790, 2025, doi: 10.1016/j.compbiomed.2025.109790.
I. Pacal, M. Alaftekin, and F. D. Zengul, “Enhancing Skin Cancer Diagnosis Using Swin Transformer with Hybrid Shifted
Window-Based Multi-head Self-attention and SwiGLU-Based MLP,” J. Imaging Informatics Med., no. 0123456789,
, doi: 10.1007/s10278-024-01140-8.
M. Arabboev, S. Begmatov, M. Rikhsivoev, S. Bobojanov, G. Tangriberganov, and J. Uraimov, “Deep learningbased
skin cancer classification: a comparative study of pre-trained CNN models,” in 30th International Conference
Information Society and University Studies - IVUS 2025, 2025.
M. Hajiarbabi, “Skin cancer detection using multi-scale deep learning and transfer learning,” J. Med. Artif. Intell., vol. 6,
no. 6, pp. 1–9, 2023, doi: 10.21037/jmai-23-67.
J. B. Thomas, K. V. Shihabudheen, S. M. Sulthan, and A. Al-Jumaily, “Deep Feature Meta-Learners Ensemble Models
for COVID-19 CT Scan Classification,” Electron., vol. 12, no. 3, 2023, doi: 10.3390/electronics12030684.
L. Sun et al., “GACEMV: An ensemble learning framework for constructing COVID-19 diagnosis and prognosis
models,” Biomed. Signal Process. Control, vol. 94, no. April, p. 106305, 2024, doi: 10.1016/j.bspc.2024.106305.
E. Erdem and T. Aydin, “Detection of Pneumonia with a Novel CNN-based Approach,” Sak. Univ. J. Comput. Inf. Sci.,
vol. 4, no. 1, pp. 26–34, 2021, doi: 10.35377/saucis.04.01.787030.
A. K. Mudiyanselage, “Deep Learning-Based Diagnosis of Pneumonia Using Convolutional Neural Networks,” J.
Comput. Mech. Manag., vol. 3, no. 3, pp. 14–21, 2024, doi: 10.57159/gadl.jcmm.3.3.240126.14.
M. Arabboev and S. Begmatov, “Deep learning-based pneumonia detection from chest X-ray images using a
convolutional neural network,” Mod. Innov. Syst. Technol., vol. 5, no. 3, pp. 1018–1026, 2025.
G. Ayana, H. Barki, and S. Choe, “Pathological Insights: Enhanced Vision Transformers for the Early Detection of
Colorectal Cancer,” Cancers (Basel)., vol. 16, no. 7, p. 1441, Apr. 2024, doi: 10.3390/cancers16071441.
A. Santone, M. Cesarelli, and F. Mercaldo, “A Method for Polyp Segmentation Through U-Net Network,” Bioengineering,
vol. 12, no. 3, pp. 1–19, 2025, doi: 10.3390/bioengineering12030236.
D. Jha et al., “Kvasir-SEG: A Segmented Polyp Dataset,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes
Artif. Intell. Lect. Notes Bioinformatics), vol. 11962 LNCS, no. April 2020, pp. 451–462, 2020, doi: 10.1007/978-3-030-
-2_37.
D. Jha et al., “A Comprehensive Study on Colorectal Polyp Segmentation with ResUNet++, Conditional Random Field
and Test-Time Augmentation,” IEEE J. Biomed. Heal. Informatics, vol. 25, no. 6, pp. 2029–2040, 2021, doi: 10.1109/
JBHI.2021.3049304.
G. Yue, W. Han, S. Li, T. Zhou, J. Lv, and T. Wang, “Automated polyp segmentation in colonoscopy images via deep
network with lesion-aware feature selection and refinement,” Biomed. Signal Process. Control, vol. 78, no. November
, p. 103846, 2022, doi: 10.1016/j.bspc.2022.103846.
M. Nouman Noor, M. Nazir, S. A. Khan, I. Ashraf, and O.-Y. Song, “Localization and Classification of Gastrointestinal
Tract Disorders Using Explainable AI from Endoscopic Images,” Appl. Sci., vol. 13, no. 15, p. 9031, Aug. 2023, doi:
3390/app13159031.
R. Nachmani, I. Nidal, D. Robinson, M. Yassin, and D. Abookasis, “Segmentation of polyps based on pyramid vision
transformers and residual block for real-time endoscopy imaging,” J. Pathol. Inform., vol. 14, no. November 2022, p.
, 2023, doi: 10.1016/j.jpi.2023.100197.
T. Shen and X. Li, “Automatic polyp image segmentation and cancer prediction based on deep learning,” Front. Oncol.,
vol. 12, no. January, pp. 1–12, 2023, doi: 10.3389/fonc.2022.1087438.
D. Bhattacharya, D. Eggert, C. Betz, and A. Schlaefer, “Squeeze and multi-context attention for polyp segmentation,”
Int. J. Imaging Syst. Technol., vol. 33, no. 1, pp. 123–142, Jan. 2023, doi: 10.1002/ima.22795.
S. Pan, X. Liu, N. Xie, and Y. Chong, “EG-TransUNet: a transformer-based U-Net with enhanced and guided models
for biomedical image segmentation,” BMC Bioinformatics, vol. 24, no. 1, pp. 1–22, 2023, doi: 10.1186/s12859-023-
-1.
D. He et al., “Dual-guided network for endoscopic image segmentation with region and boundary cues,” Biomed.
Signal Process. Control, vol. 91, no. January, p. 106059, 2024, doi: 10.1016/j.bspc.2024.106059.
M. F. Ahamed, M. R. Islam, M. Nahiduzzaman, M. E. H. Chowdhury, A. Alqahtani, and M. Murugappan, “Automated
Colorectal Polyps Detection from Endoscopic Images using MultiResUNet Framework with Attention Guided
Segmentation,” Human-Centric Intell. Syst., vol. 4, no. 2, pp. 299–315, 2024, doi: 10.1007/s44230-024-00067-1.
W. Dong, B. Du, and Y. Xu, “Shape-intensity-guided U-net for medical image segmentation,” Neurocomputing, vol.
, no. March, 2024, doi: 10.1016/j.neucom.2024.128534.
N. Tirpude, T. Diwan, and M. Dhabu, “Meta-Transformers: A Hybrid Approach for Medical Image Segmentation
with U-Net and Meta-Learning,” IEEE Access, vol. 13, no. February, pp. 64822–64831, 2025, doi: 10.1109/
ACCESS.2025.3559446.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 MUHANDISLIK VA IQTISODIYOT
This work is licensed under a Creative Commons Attribution 4.0 International License.