AUT Journal of Electrical Engineering

AUT Journal of Electrical Engineering

Enhancing Colonoscopy-Images Segmentation using Laplacian-Former as Efficient Transformer Block

Document Type : Research Article

Authors
Jabraeil Ranjbar
Hossein Ebrahimnezhad
Mohammad Hossein Sedaaghi
Faculty of Electrical Engineering, Sahand University of Technology, Tabriz, Iran
10.22060/eej.2025.23498.5619
Abstract
This research is dedicated to enhancing medical image segmentation through the innovative Laplacian-Former algorithm, which introduces novel and responsive techniques in image processing. Given the crucial role of precise image segmentation in disease diagnosis, challenges such as achieving high accuracy, sensitivity, and reliability persist in existing methods, necessitating advanced approaches. The method proposed based on Laplacian-Former is designed to enhance the accuracy and efficiency of medical image segmentation. The algorithm addresses limitations in current segmentation techniques, providing robust solutions for medical imaging tasks. Through a detailed analysis of experimental outcomes and a comparative assessment of Laplacian-Former against existing techniques, it is evident that this approach not only contributes to segmentation accuracy but also enhances the sensitivity and reliability of medical images substantially. This study introduces a flexible architecture adaptable to various segmentation tasks, with a novel focus on polyp segmentation in colonoscopy images. The results obtained suggest that employing Laplacian-Former as an innovative technique in medical image segmentation can lead to enhanced disease detection accuracy and performance. Evaluation of the widely recognized Kvasir dataset demonstrates outstanding results, achieving top performance metrics, even with a small training dataset. Data augmentation methods were applied to improve training, further enhancing model proficiency. This investigation vitrines the algorithm's value in advancing medical image processing and research. The Laplacian-Former holds significant promise for future medical imaging advancements and improved disease diagnosis.
Keywords
Medical Image Segmentation
Laplacian-Former Algorithm
Polyp segmentation
Colonoscopy Images
Deep Learning
Subjects
  1. Azad, R. et al. Laplacian-Former: Overcoming the Limitations of Vision Transformers in Local Texture Detection. In: Greenspan, H., et al. Medical Image Computing and Computer Assisted Intervention – MICCAI 2023. MICCAI 2023. Lecture Notes in Computer Science, vol 14222. Springer, Cham. https://doi.org/10.1007/978-3-031-43898-1_70 (2023).
  2. Wang, P., Zheng, W., Chen, T., Wang, Z.: Anti-over smoothing in deep vision transformers via the Fourier domain analysis: From theory to practice. In: International Conference on Learning Representations (2022), https://openreview.net/forum? id=O476oWmiNNp (2022).
  3. Helen M. Mohan, Julie M.L. Sijmons, Jack V. Maida, Kate Walker, Angela Kuryba, Ingvar Syk, Lene H. Iversen, Alexander Hariot, Clifford Y. Ko, Pieter J. Tanis, Rob A.E.M. Tollenaar, Nicholas Avellaneda, Philip Smart, Identifying a common data dictionary across colorectal cancer outcome registries: A mapping exercise to identify opportunities for data dictionary harmonization, European Journal of Surgical Oncology, Volume 50, Issue 2, 2024, 107937, ISSN 0748-7983, https://www.sciencedirect.com/science/article/pii/S0748798323015755 (2024).
  4. American Cancer Society. Colorectal cancer early detection, diagnosis, and staging. Retrieved from https://www.cancer.org/cancer/colon-rectal-cancer/detection-diagnosis-staging/detection.html (2021)
  5. Shaukat, A. et al. ACG Clinical Guidelines: Colorectal Cancer Screening 2021. The American Journal of Gastroenterology 116(3), 458-479. https://doi.org/10.14309/ajg.0000000000001122 (2021).
  6. Li, Y. Liu, X. Liu, and Y. Qi, "Fusing Transformer and FCN for Polyp Segmentation," 2023 7th Asian Conference on Artificial Intelligence Technology (ACAIT), Jiaxing, China, 2023, pp. 623-628, doi: 10.1109/ACAIT60137.2023.10528540 (2023).
  7. Tharwat, M., Sakr, N. A., El-Sappagh, S., Soliman, H., Kwak, K., & Elmogy, M. Colon Cancer Diagnosis Based on Machine Learning and Deep Learning: Modalities and Analysis Techniques. Sensors, 22(23), 9250. https://doi.org/10.3390/s22239250 (2022). 21
  8. K. Rai, D. Singh and A. Shukla, "Polyp Detection Using U-Net Neural Network Based Algorithm," 2024 2nd International Conference on Disruptive Technologies (ICDT), Greater Noida, India, 2024, pp. 1367-1373, doi: 10.1109/ICDT61202.2024.10488975 (2024).
  9. Li et al., "Colon polyp segmentation method based on global contextual information and U-shaped network," 2024 5th International Conference on Computer Vision, Image and Deep Learning (CVIDL), Zhuhai, China, 2024, pp. 892-896, doi: 10.1109/CVIDL62147.2024.10603834 (2024).
  10. Wang et al., "An Efficient Multi-Task Synergetic Network for Polyp Segmentation and Classification," in IEEE Journal of Biomedical and Health Informatics, vol. 28, no. 3, pp. 1228-1239, March 2024, doi: 10.1109/JBHI.2023.3273728 (2024).
  11. Panagiotidou, A. Kakasis and I. Bensenousi, "ONCO-AICO: An AI-Based Educational Tool for Polyp Detection from Colonoscopy," 2024 5th International Conference in Electronic Engineering, Information Technology & Education (EEITE), Chania, Greece, 2024, pp. 1-2, doi: 10.1109/EEITE61750.2024.10654445 (2024).
  12. Ronneberger, O., Fischer, P., & Brox, T. U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention - MICCAI (pp. 234-241). Springer International Publishing https://doi.org/10.1007/978-3-319-24574-4_28. (2015).
  13. Long, J., Shelhamer, E., & Darrell, T. Fully convolutional networks for semantic segmentation. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 3431-3440). https://doi.org/10.1109/CVPR.2015.7298965 (2015).
  14. Zhou, Z., Rahman Siddiquee, M. M., Tajbakhsh, N., & Liang, J. UNet++: A Nested U-Net Architecture for Medical Image Segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (pp. 3-11). Springer International Publishing. https://doi.org/10.1007/978-3-030-00889-5_1 (2018).
  15. Seo, et al. Modified U-Net (mU-Net) With Incorporation of Object-Dependent High-Level Features for Improved Liver and Liver-Tumor Segmentation in CT Images. IEEE Transactions on Medical Imaging, 39, 5, pp. 1316-1325. doi: https://doi.org/10.1109/TMI.2019.2948320 (2020).
  16. Jha, D. et al. ResUNet++: An Advanced Architecture for Medical Image Segmentation. IEEE International Symposium on Multimedia (ISM) (pp. 225-2255). https://doi.org/10.1109/ISM46123.2019.00049 (2019).
  17. Li, X. et al. H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes. IEEE Transactions on Medical Imaging, 37(12), 2663-2674. https://doi.org/10.1109/TMI.2018.2845918 (2018).
  18. Chen, L.-C., Papandreou, G., Schroff, F., & Adam, H. Rethinking Atrous Convolution for Semantic Image Segmentation. CoRR, abs/1706.05587. Preprint at https://arxiv.org/abs/1706.05587 (2017).
  19. Bernal, et al. Comparative Validation of Polyp Detection Methods in Video Colonoscopy: Results From the MICCAI 2015 Endoscopic Vision Challenge. IEEE Transactions on Medical Imaging, 36, 6, pp. 1231-1249. doi: https://doi.org/10.1109/TMI.2017.2664042 (2017).
  20. Buslaev, A., et al. Albumentations: Fast and Flexible Image Augmentations. Information, 11(2), 125. doi: https://doi.org/10.3390/info11020125">10.3390/info11020125 (2020).
  21. Geirhos, R., Rubisch, P., Michaelis, C., Bethge, M., Wichmann, F.A., Brendel, W.: Imagenet-trained cnns are biased towards texture; increasing shape bias improves accuracy and robustness. In: International Conference on Learning Representations.
  22. Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., Guo, B.: Swin transformer: Hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 10012–10022 (2021).
  23. Huang, X., Deng, Z., Li, D., Yuan, X., Fu, Y.: Missformer: An effective transformer for 2d medical image segmentation. IEEE Transactions on Medical Imaging 1–1 (2022). https://doi.org/10.1109/TMI.2022.3230943.
  24. Cao, H., Wang, Y., Chen, J., Jiang, D., Zhang, X., Tian, Q., Wang, M.: Swin-unet: Unet-like pure transformer for medical image segmentation. In: Computer Vision– ECCV 2022 Workshops: Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part III. pp. 205–218. Springer (2023).
  25. Wang, P., Zheng, W., Chen, T., Wang, Z.: Anti-overs moothing in deep vision transformers via the Fourier domain analysis: From theory to practice. In: International Conference on Learning Representations (2022), https://openreview.net/forum? id=O476oWmiNNp.
  26. Shen, Z., Zhang, M., Zhao, H., Yi, S., Li, H.: Efficient attention: Attention with linear complexities. In: Proceedings of the IEEE/CVF winter conference on applications of computer vision. pp. 3531–3539 (2021).
  27. Fan, DP. et al. PraNet: Parallel Reverse Attention Network for Polyp Segmentation. Medical Image Computing and Computer Assisted Intervention. Lecture Notes in Computer Science, 12266. Springer, Cham. https://doi.org/10.1007/978-3-030-59725- 2_26 (2020).
  28. Duc, N. T., Oanh, N. T., Thuy, N. T., Triet, T. M., & Dinh, V. S. ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation. IEEE Access, 10, 80575-80586. https://doi.org/10.1109/ACCESS.2022.3195241 (2022).
  29. Sun, K., Xiao, B., Liu, D., & Wang, J. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 5693-5703 (2019).
  30. Sun, K., et al. High-Resolution Representations for Labeling Pixels and Regions. CoRR, abs/1904.04514. Preprint at http://arxiv.org/abs/1904.04514 (2019).
  31. Diakogiannis, F. I., Waldner, F., Caccetta, P., & Wu, C. ResUNet-a: A deep learning framework for semantic segmentation of remotely sensed data. ISPRS Journal of Photogrammetry and Remote Sensing, 162, 94-114. https://doi.org/10.1016/j.isprsjprs.2020.01.013 (2020).
  32. Liao, T. Y., et al. HarDNet-DFUS: An Enhanced Harmonically-Connected Network for Diabetic Foot Ulcer Image Segmentation and Colonoscopy Polyp Segmentation. Preprint at https://arxiv.org/abs/2209.07313 (2022).
  33. Duc, N. T., Oanh, N. T., Thuy, N. T., Triet, T. M., & Dinh, V. S. ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation. IEEE Access, 10, 80575-80586. https://doi.org/10.1109/ACCESS.2022.3195241 (2022).
  34. Srivastava, A., et al. MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation. IEEE Journal of Biomedical and Health Informatics, 26(5), 2252-2263. https://doi.org/10.1109/JBHI.2021.3138024 (2022).
  35. Deng, J., et al. ImageNet: A Large-Scale Hierarchical Image Database. IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2009) (pp. 248-255). https://doi.org/10.1109/CVPR.2009.5206848 (2009).
  36. Sanderson, E., & Matuszewski, B. J. FCN-Transformer Feature Fusion for Polyp Segmentation. Medical Image Understanding and Analysis (pp. 892-907). Springer International Publishing, https://doi.org/10.1007/978-3-031-12053-4_65 (2022).
  37. Wang, J., et al. Stepwise Feature Fusion: Local Guides Global. Preprint at https://arxiv.org/abs/2203.03635 (2022).
  38. Dosovitskiy, A., et al. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. Preprint at https://arxiv.org/abs/2010.11929 (2020).
  39. Tan, M., & Le, Q. EfficientNetV2: Smaller Models and Faster Training. Proceedings of the 38th International Conference on Machine Learning (pp. 10096-10106).
  40. Deng, J., et al. ImageNet: A Large-Scale Hierarchical Image Database. IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2009) (pp. 248-255). https://doi.org/10.1109/CVPR.2009.5206848 (2009).
  41. Krizhevsky, A. & Hinton, G. Learning multiple layers of features from tiny images. Technical Report, 2009. Retrieved from https://www.cs.toronto.edu/~kriz/learningfeatures-2009-TR.pdf (2009).
  42. Tang, Y., Han, K., Xu, C., Xiao, A., Deng, Y., Xu, C., Wang, Y.: Augmented shortcuts for vision transformers. Advances in Neural Information Processing Systems 34, 15316–15327 (2021).
  43. Gu, J., Kwon, H., Wang, D., Ye, W., Li, M., Chen, Y.H., Lai, L., Chandra, V., Pan, D.Z.: Multi-scale high-resolution vision transformer for semantic segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 12094–12103 (2022).
  44. Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J.M., Luo, P.: Segformer: Simple and efficient design for semantic segmentation with transformers. Advances in Neural Information Processing Systems 34, 12077–12090 (2021).
  45. Buslaev, A., et al. Albumentations: Fast and Flexible Image Augmentations. Information, 11(2), 125. doi: https://doi.org/10.3390/info11020125">10.3390/info11020125 (2020).
  46. Perruchet, P.; Peereman, R. (2004). "The exploitation of distributional information in syllable processing".  Neurolinguistics. 17(2–3): 97–119.
  47. Cohen, W. W., Ravikumar, P., & Fienberg, S. E. (2003). A comparison of string distance metrics for name-matching tasks. In IIWeb (Vol. 3, No. 73, pp. 73-78).
  48. M. Weiss, N. Indurkhya, T. Zhang, "Fundamentals of Predictive Text Mining", Springer Science & Business Media, 2010.
  49. McClave, J. T., Sincich, T., & Mendenhall, W. (2019). Statistics (13th ed.). Pearson.
  50. Jha, D. et al. Kvasir-SEG: A Segmented Polyp Dataset. MultiMedia Modeling. MMM 2020. Lecture Notes in Computer Science, 11962. Springer, Cham. https://doi.org/10.1007/978-3-030-37734-2_37 (2020).
  51. Amin Golzari Oskouei, Nasim Abdolmaleki, Asgarali Bouyer, Bahman Arasteh, Kimia Shirini, Efficient super pixel-based brain MRI segmentation using multi-scale morphological gradient reconstruction and quantum clustering, Biomedical Signal Processing and Control, Volume 100, Part B, 2025, 107063, ISSN 1746-8094, https://doi.org/10.1016/j.bspc.2024.107063.
  52. Akan, T., Oskouei, A.G., Alp, S. et al. Brain magnetic resonance image (MRI) segmentation using multimodal optimization. Multimed Tools Appl (2024). https://doi.org/10.1007/s11042-024-19725-4.
  53. Khiarak, J.N., Oskouei, A.G., Nasab, S.S. et al. KartalOl: a new deep neural network framework based on transfer learning for iris segmentation and localization task—new dataset for iris segmentation. Iran J Comput Sci6, 307–319 (2023). https://doi.org/10.1007/s42044-023-00141-0.
  54. Oskouei, A. G., Balafar, M. A., & Akan, T. (2023). A brain MRI segmentation method using feature weighting and a combination of efficient visual features. In Applied computer vision and soft computing with interpretable AI (1st ed., pp. 20). Chapman and Hall/CRC. https://doi.org/10.4324/9781003359456.
  55. Chen, J., Lu, Y., Yu, Q., Luo, X., Adeli, E., Wang, Y., Lu, L., Yuille, A.L., & Zhou, Y. (2021). TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation. ArXiv, abs/2102.04306.
  56. Chen, LC., Zhu, Y., Papandreou, G., Schroff, F., Adam, H. (2018). Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds) Computer Vision – ECCV 2018. ECCV 2018. Lecture Notes in Computer Science(), vol 11211. Springer, Cham. https://doi.org/10.1007/978-3-030-01234-2_49.
  57. Heidari, M., Kazerouni, A., Soltany, M., Azad, R., Aghdam, E.K., Cohen-Adad, J., & Merhof, D. (2023). HiFormer: Hierarchical Multi-scale Representations Using Transformers for Medical Image Segmentation. In Proceedings - 2023 IEEE Winter Conference on Applications of Computer Vision, WACV 2023(pp. 6191-6201). Waikoloa, HI, USA: Institute of Electrical and Electronics Engineers Inc.
  58. Azad, Reza & Heidari, Moein & Shariatnia, Moein & Khodapanah Aghdam, Ehsan & Karimijafarbigloo, Sanaz & Adeli, Ehsan & Merhof, Dorit. (2022). TransDeepLab: Convolution-Free Transformer-Based DeepLab v3+ for Medical Image Segmentation. 10.1007/978-3-031-16919-9_9.
  59. Li, Feng & Huang, Zetao & Zhou, Lu & Chen, Yuyang & Tang, Shiqing & Ding, Pengchao & Peng, Haixia & Chu, Yimin. (2024). Improved dual-aggregation polyp segmentation network combining a pyramid vision transformer with a fully convolutional network. Biomedical Optics Express. 15. 10.1364/BOE.510908.
  60. Bernal, J., et al. WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians. Computerized Medical Imaging and Graphics, 43, 99-111. doi: https://doi.org/10.1016/j.compmedimag.2015.02.007 (2015).
  61. Bernal, J., et al. WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians. Computerized Medical Imaging and Graphics, 43, 99-111. doi: https://doi.org/10.1016/j.compmedimag.2015.02.007 (2015).