Mamun Ahmed*, Shovan Bhowmik, Md. Imrul Kayes, Rashida Feroz Prome and Maria Noor
Abstract: Computer Vision and Deep Learning algorithms have assisted the disease diagnosis process by extracting helpful information from related data. Even though the prevalence of colorectal cancer has been steadily rising, there have been few attempts to develop a reliable and specialized computational approach for early diagnosis. One technique to prevent this disease is to recognize the polyps in the colon walls. Previously, large-scale image identification algorithms, Conventional Data Mining approaches, and Deep Neural Networks were trained using RGB and Monochrome Scale images. However, they were based on highly resourced, pretrained Convolutional Neural Network (CNN) models. As CNNs are widely used in medical image classification and segmentation, we have devised a lightweight and simple CNN architecture dedicated to making faster training only on colon polyp images. This study utilised a three-layered CNN to distinguish two kinds of colon polyps that cause colorectal cancer. On a benchmark dataset, this network correctly classified 97.60% ‘Hyperplastic’ and ‘Adenomatous’ polyps with a minimum training time of approximately 32 minutes which is considerably lowered compared to conventional models.
Keywords: Colorectal cancer, colon polyps, convolutional neural network, training time, HOG detector, support vector machine
References
Brownlee, J. (2021). A Gentle Introduction to Ensemble Learning Algorithms. Machine Learning Mastery. learning-algorithms/
Banik, D., Roy, K., Bhattacharjee, D., Nasipuri, M., & Krejcar, O. (2020). Polyp-Net: A multimodel fusion network for polyp segmentation. IEEE Transactions on Instrumentation and Measurement, 70, 1–12.
Bhandari, A. (2020). Image augmentation on the fly using Keras Image Data Generator. Analytics Vidya.
Bray, F., Ferlay, J., Soerjomataram, I., Siegel, R. L., Torre, L. A., & Jemal, A. (2018). Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians, 68(6), 394–424.
El Khatib, A., Werghi, N., & Al-Ahmad, H. (2015, August). Automatic polyp detection: A comparative study. In 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (pp. 2669–2672). IEEE.
European Colorectal Cancer Screening Guidelines Working Group. (2013). European guidelines for quality assurance in colorectal cancer screening and diagnosis: overview and introduction to the full supplement publication. Endoscopy, 45(01), 51–59.
Feature Selection Techniques in Machine Learning with Python. 2021. https://towardsdatascience.com/feature-selectiontechniques-in-machine-learningwith-python-f24e7da3f36e
Freedman, D., Blau, Y., Katzir, L., Aides, A., Shimshoni, I., Veikherman, D., & Rivlin, E. (2020). Detecting deficient coverage in colonoscopies. IEEE Transactions on Medical Imaging, 39(11), 3451–3462.
Gessl, I., Waldmann, E., Penz, D., Majcher, B., Dokladanska, A., Hinterberger, A., & Ferlitsch, M. (2019). Evaluation of adenomas per colonoscopy and adenomas per positive participant as new quality parameters in screening colonoscopy. Gastrointestinal Endoscopy, 89(3), 496–502.
Jha, D., Smedsrud, P. H., Johansen, D., de Lange, T., Johansen, H. D., Halvorsen, P., & Riegler, M. A. (2021). A comprehensive study on colorectal polyp segmentation with ResUNet++, conditional random field and test-time augmentation. IEEE Journal of Biomedical and Health Informatics, 25(6), 2029–2040.
Kammoun, A., & AlouiniFellow, M. S. (2021). On the precise error analysis of support vector machines. IEEE Open Journal of Signal Processing, 2, 99–118.
Kang, J., & Gwak, J. (2019). Ensemble of instance segmentation models for polyp segmentation in colonoscopy images. IEEE Access, 7, 26440–26447.
American Cancer Society. (2022). Key statistics for colorectal cancer. (2021). https://www.cancer.org/cancer/colon-rectal-cancer/about/key-statistics.html
Li, K., Fathan, M. I., Patel, K., Zhang, T., Zhong, C., Bansal, A., & Wang, G. (2021). Colonoscopy polyp detection and classification: Dataset creation and comparative evaluations.arXiv preprint arXiv:2104.10824.
Liu, B., Guo, W., Chen, X., Gao, K., Zuo, X., Wang, R., & Yu, A. (2020). Morphological attribute profile cube and deep random forest for small sample classification of hyperspectral image. IEEE Access, 8, 117096–117108.
Nguyen, N. Q., Vo, D. M., & Lee, S. W. (2020). Contour-aware polyp segmentation in colonoscopy images using detailed upsamling encoder-decoder networks. IEEE Access, 8, 99495–99508.
Nisha, J. S., Gopi, V. P., & Palanisamy, P. (2022). Automated colorectal polyp detection based on image enhancement and dual-path CNN architecture. Biomedical Signal Processing and Control, 73, 103465.
Official website of Keras API. (2021). https://keras.io/api/preprocessing/image/ Paper, D., & Paper, D. (2020). Scikit-learn classifier tuning from simple training sets. Hands-on Scikit-Learn for Machine Learning Applications: Data Science Fundamentals with Python, 137–163.
Patel, K., Li, K., Tao, K., Wang, Q., Bansal, A., Rastogi, A., & Wang, G. (2020). A comparative study on polyp classification using convolutional neural networks. PloS one, 15(7), e0236452.
Poudel, S., Kim, Y. J., Vo, D. M., & Lee, S. W. (2020). Colorectal disease classification using efficiently scaled dilation in convolutional neural network. IEEE Access, 8, 99227–99238.
Pozdeev, A. A., Obukhova, N. A., & Motyko, A. A. (2019, January). Automatic analysis of endoscopic images for polyps detection and segmentation. In 2019 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus) (pp. 1216–1220). IEEE.
Qadir, H. A., Balasingham, I., Solhusvik, J., Bergsland, J., Aabakken, L., & Shin, Y. (2019). Improving automatic polyp detection using cnn by exploiting temporal dependency in colonoscopy video. IEEE Journal of Biomedical and Health Informatics, 24(1), 180–193.
Qian, Z., Lv, Y., Lv, D., Gu, H., Wang, K., Zhang, W., & Gupta, M. M. (2020). A new approach to polyp detection by preprocessing of images and enhanced faster RCNN. IEEE Sensors Journal, 21(10), 11374–11381.
Rani, N., Verma, R., & Jindal, A. (2022). Polyp Detection Using Deep Neural Networks. Handbook of Intelligent Computing and Optimization for Sustainable Development, 801–814.
Ribeiro, E., Uhl, A., & Häfner, M. (2016, June). Colonic polyp classification with convolutional neural networks. In 2016 IEEE 29th International Symposium on Computer-Based Medical Systems (CBMS) (pp. 253–258). IEEE.
Shah, S. A., Brown, P., Gimeno, H., Lin, J. P., & McClelland, V. M. (2020). Application of machine learning using decision trees for prognosis of deep brain stimulation of globus pallidus internus for children with dystonia. Frontiers in neurology, 825.
Shalini, M., & Radhika, S. (2020, February). Machine learning techniques for prediction from various breastcancer datasets. In 2020 Sixth International Conference on Bio Signals, Images, and Instrumentation (ICBSII) (pp. 1–5). IEEE. Tjaden, J. M., Hause, J. A., Berger, D., Duveneck, S. K., Jakate, S. M., Orkin, B. A., … & Melson, J. E. (2018). Adenoma detection rate metrics in colorectal cancer surveillance colonoscopy. Surgical endoscopy, 32(7), 3108–3113.
Xi, Y., & Xu, P. (2021). Global colorectal cancer burden in 2020 and projections to 2040. Translational Oncology, 14(10), 101–174.
Yamashita, R., Nishio, M., Do, R. K. G., & Togashi, K. (2018). Convolutional neural networks: an overview and application in radiology. Insights into Imaging, 9(4), 611–629.
Yang, X., Wei, Q., Zhang, C., Zhou, K., Kong, L., & Jiang, W. (2020). Colon polyp detection and segmentation based on improved MRCNN. IEEE Transactions on Instrumentation and Measurement, 70, 1–10.