Feature extraction and Classification of colonoscopy lesions based on the HCBA framework

Subject Areas : Multimedia Processing, Communications Systems, Intelligent Systems

Vahid Bayrami Rad ¹ , Mahdi Mazinani ^{2 *} , Mitra Mirzarezaee ³

1 - Ph.D. Student, Department of Computer Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
2 - Associate professor, Department of Electrical and Electronic Engineering, Shahr-e-Qods Branch, Islamic Azad University, Tehran, Iran
3 - Associate professor, Department of Computer Engineering, Science and Research Branch, Islamic Azad University, Tehran Iran

Received: 2024-09-24 Accepted : 2024-11-05 Published : 2025-01-27

Keywords: Colonoscopy, Colon Lesions, Deep Learning, Image Processing, Classification,

Abstract :

Abstract

Introduction: Recent years have witnessed expansion of colon-related diseases expanding due to the expansion of modern lifestyles. Fortunately, early diagnosis of these diseases can significantly facilitate the treatment and control of these diseases. A novel deep learning framework, HCBA, was introduced to classify colonoscopy lesions. The framework consists of several methods such as hierarchical clustering, a bag of features, and an auto-encoder.

Method: Feature extraction was done using BOF and autoencoder, where hierarchical clustering and colonoscopy lesion classification were performed using DBN.

Results: The proposed method had a high performance in classifying colonoscopy images. The method exhibited the highest mean results in HCBA-DBN, including 93.0 in ACC, 87.1 in Pre, 87.6 in Rec, and 94.8 in Sen.

Discussion: The extraction of features was based on a set of features and hierarchical clustering was optimized. The separation of the similarity measurement between the samples within the cluster was based on the standard deviation. In addition, deep belief learning was used.

 

References:

1. Jheng, Y.-C., et al., A novel machine learning-based algorithm to identify and classify lesions and anatomical landmarks in colonoscopy images. Surgical Endoscopy, 2022. 36: p. 640-650.
2. Zhang, R., et al., Automatic detection and classification of colorectal polyps by transferring low-level CNN features from nonmedical domain. IEEE journal of biomedical and health informatics, 2016. 21(1): p. 41-47.
3. ELKarazle, K., et al., Detection of colorectal polyps from colonoscopy using machine learning: A survey on modern techniques. Sensors, 2023. 23(3): p. 1225.
4. Wimmer, G., et al., Fisher encoding of convolutional neural network features for endoscopic image classification. Journal of Medical Imaging, 2018. 5(3): p. 034504-034504.
5. Cherubini, A., A novel AI device for real-time optical characterization of colorectal polyps.
6. Gong, E.J., et al., No-code platform-based deep-learning models for prediction of colorectal polyp histology from white-light endoscopy images: development and performance verification. Journal of Personalized Medicine, 2022. 12(6): p. 963.
7. Biffi, C., et al., A novel AI device for real-time optical characterization of colorectal polyps. NPJ digital medicine, 2022. 5(1): p. 84.
8. Zhou, J., et al., Feature fusion for diagnosis of atypical hepatocellular carcinoma in contrast-enhanced ultrasound. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2021. 69(1): p. 114-123.
9. Chen, C., et al., Domain knowledge powered deep learning for breast cancer diagnosis based on contrast-enhanced ultrasound videos. IEEE Transactions on Medical Imaging, 2021. 40(9): p. 2439-2451.
10. Guo, Z., et al. Automatic polyp recognition from colonoscopy images based on bag of visual words. in 2017 IEEE 8th International Conference on Awareness Science and Technology (iCAST). 2017. IEEE.
11. Saluja, K., et al. Efficient bag of deep visual words based features to classify CRC images for colorectal tumor diagnosis. in 2022 2nd international conference on advance computing and innovative technologies in engineering (ICACITE). 2022. IEEE.
12. Nawarathna, R., et al., Abnormal image detection in endoscopy videos using a filter bank and local binary patterns. Neurocomputing, 2014. 144: p. 70-91.
13. Huang, H. and Y. Ma, A hybrid clustering approach for bag-of-words image categorization. Mathematical Problems in Engineering, 2019. 2019: p. 1-11.
14. Stasiowski, M.J., et al., Adequacy of anesthesia guidance for colonoscopy procedures. Pharmaceuticals, 2021. 14(5): p. 464.
15. Wang, D., et al., Afp-mask: Anchor-free polyp instance segmentation in colonoscopy. IEEE Journal of Biomedical and Health Informatics, 2022. 26(7): p. 2995-3006.
16. Tian, Y., et al., Detecting, localizing and classifying polyps from colonoscopy videos using deep learning, in Deep Learning for Medical Image Analysis. 2024, Elsevier. p. 425-450.
17. Hossain, M.S., et al., Deeppoly: deep learning based polyps segmentation and classification for autonomous colonoscopy examination. IEEE Access, 2023.
18. Ma, C., et al. A Real-Time Polyp Detection Framework for Colonoscopy Video. in Chinese Conference on Pattern Recognition and Computer Vision (PRCV). 2022. Springer.
19. Nisha, J., V.P. Gopi, and P. Palanisamy, Automated colorectal polyp detection based on image enhancement and dual-path CNN architecture. Biomedical Signal Processing and Control, 2022. 73: p. 103465.
20. Carrinho, P. and G. Falcao, Highly accurate and fast YOLOv4-Based polyp detection. Expert Systems with Applications, 2023. 232: p. 120834.
21. Krenzer, A., et al., A real-time polyp-detection system with clinical application in colonoscopy using deep convolutional neural networks. Journal of imaging, 2023. 9(2): p. 26.