CNN AND HYBRID NEURAL-SVM TECHNIQUE FOR BIOMEDICAL IMAGE ENHANCEMENT
Keywords:
Convolution neural networks (CNN), support vector machine, magnetic resonance imaging (MRI), denoising, enhancementAbstract
This paper presents an advanced approach to increase the quality of biomedical images based upon a hybrid architecture consisting of a Convolutional neural network (CNN). In the proposed method, we have applied a combination of Genetic algorithm and wavelet as a first step to improve the quality of biomedical images such as MRI, CT, X-ray, PET, and Ultrasound images from different body organs. Noisy and poor quality biomedical images have been a big challenge for doctors during diagnosis for years. Therefore, a CNN approach has been used for image denoising. CNN has a large modeling capacity and remarkable advances in network training and design. Additionally, CNN can be embedded as a modular part of traditional methods, which can be used with greater efficiency. As in CNN, the new image formed at each layer due to convolution is passed to the next layer, so it provides better images than the neural networks and is best suited for structured numerical data. Experimental results show that the proposed technique using Convolutional neural networks provides better quantitative results in terms of structured similarity index (SSIM), mean squared error (MSE), standard deviation, accuracy, sensitivity and beta coefficient. Finally, the comparison of Convolutional neural networks and the hybrid Neural-SVM method is done that indicates the best results of CNN in terms of less computational time.
References
Abreu, E., Lightstone, M., Mitra, S.K., and Arakawa, K. (1996). A New efficient approach for the removal of impulse noise from highly corrupted images. IEEE Trans. on Image Proc., 5:13, 1,012.
Agostinelli, F., Anderson, M.R., and Lee, H. (2013). Adaptive multi-column deep neural networks with application to robust image denoising. Proc. Adv. Neural Inf. Process. Syst., p. 1,493-1,501.
Ahmmed, R., Swakshar, A.S., Hossain, F., and Rafiq, A. (2017). Classification of tumors and its stages in brain mri using support vector machine and artificial neural network. International Conference on Electrical, Computer and Communication Engineering (ECCE): February 16-18, 2017: Chittagong, Bangladesh, p. 229-234.
Ali, S.A., Vathsal, S., and kishore, L. (2010). CT image denoising technique using ga aided window-based multiwavelet transformation and thresholding with the incorporation of an effective quality enhancement method. Int. J. Digital Cont. Tech. its App., 4(4):12,75.
Astola, J., Haavisto, P., and Neuovo, Y. (1990). Vector median filters. Proceeding of Image Processing, IEEE, 78:10, 678.
Beis, J. and Lowe, D. (1997). Shape indexing using approximate nearest -neighbor search in high dimensional spaces. IEEE International Conference on Computer Vision and Pattern Recognition: June, 1997: San Juan, Puerto Rico, USA, p. 6, 1,000.
Bhattacharya, P., Riechen, J., and Zolzer, U. (2018). Infrared image enhancement in maritime environment with convolutional neural networks. In: Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2018: 4:37-46
Bilcu, R.C. and Vehvilainen, M. (2007). A novel decomposition scheme for image de-noising. IEEE International Conference on Acoustics, Speech and Signal Processing: April 15-20, 2007, Hawaii, USA, p. 577-580.
Chang, Y., Hsieh, C., Chang, K., Ringgaard, M., and Lin, C. (2010). Training and testing low-degree polynomial data mappings via linear SVM. J. Mach. Learn. Res., 11:19, 1,471.
Chen, H., Zhang, Y., Kalra, M. K., Lin, F., Liao, P., Zhou, J., and Wang, G. (2017). Low-dose CT with a residual encoder-decoder convolutional neural network (REDCNN). arXiv: 1702.00288.
Choi, Y., Kim, N., Hwang, S., and Kweon, I. (2016). Thermal image enhancement using convolutional neural network. 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), p. 223-230.
Chunhui, Z., Qing bin, X., and Wei, N. (1998). Study on the noise attenuation characteristics of generalized morphological filters. In: SPIE Proceeding Medical Imaging: September 25, 1998: Wuhan, China, p. 236-39.
Descombes, X., Kruggel, F., Wollny, G., and Gertz, H.J. (2004). An object based approach for detecting small brain lesions: Application to Virchow-robin spaces. IEEE Trans. Med. Imag., 23(2):9,246.
Dong, C., Loy, C., C., He, K. and Tang, X. (2016). Image Super-resolution Using Deep Convolutional networks. IEEE Trans. Patt. Anal. Mach. Intell., 38(2):295-307.
Dufan, W., Kyungsang, K., Georges, E.F., and Quanzheng, L. (2017). A cascaded convolutional neural network for X-ray low-dose CT image denoising. arXiv:1705.04267v2 (cs).
Fukushima, K. (1980). Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Bio. Cyber., 36(4):193-202.
Gupta, B. and Tiwari, S. (2014). Brain tumor detection using curvelet transform and support vector machine. Int. J. Comp. Sci. Mobile Comp., 3(4):5,1,259.
Ishak, N.F., Logeswaran, R., and Tan, W. (2008). Artifact and noise stripping on low-field brain MRI. International J. Bio. Biomed. Eng., 2(2):59-68.
Jain, V. and Seung, S. (2009). Natural image denoising with convolutional networks. Adv. Neural Inform. Proc. Sys., p. 769-776.
Kang, E., Min, J., and Ye, J.C. (2016). A deep convolutional neural network using directional wavelets for low-dose X-ray CT reconstruction. arXiv: 1610.09736.
Kaur, L., Gupta, S., and Chauhan. (2002). Image denoising using Wavelet Thresholding. Indian Conference on Computer Vision, Graphics and Image Processing, 2002.
Karthikeyan, K., and Chandrasekhar, C. (2011). Speckle noise reduction of medical ultrasound images using bayes shrink wavelet threshold. Int. J. Comput. Appl., 22:6, 8.
Khatkar, K. and Kumar, D. (2018). A new Genetic Algorithm based technique for biomedical image Enhancement. Pertanika J. Sci. Tech., 26(4):25,1,725.
Kim, J., Lee, J.K., and Lee, K.M. (2016). Accurate image super-resolution using very deep convolutional networks. IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), IEEE. doi:10.1109/CVPR. 2016.182. p. 1,646-1,654.
Larose, D.T. (2005). Discovering knowledge in data: an introduction to data mining. Wiley-Interscience, California State University, Fullerton, p. 1,465.
Liu, P., Zhang, H., Lian, W. and Zuo, W. (2019). Multi-level wavelet convolutional neural networks. arXiv: 1907.03128(cs).
Lugina, M., Retno, N.D., and Rita, R. (2015). Brain tumor detection and classification in magnetic resonance imaging (mri) using region growing, fuzzy symmetric measure, and artificial neural network back propagation. Int. J. ICT, 1:8, 20.
Mao, X., Shen, C., and Yang, Y. (2016). Image denoising using very deep fully convolutional encoder-decoder networks with symmetric skip connections. arXiv: 1603.09056v2.
Nandpuru, B.H., Salankar, S.S., and Bora, R.V. (2014). MRI brain cancer classification using Support Vector Machine. 2014 IEEE Students’ Conference on Electrical, Electronics and Computer Science, March, 2014, Bhopal, India.
Osovsky, S. (2002). The neural networks for data Processing. Finances and Statistics, Moscow.
Padma, A. and Sukanesh, R. (2013). SVM based classification of soft tissues in brain CT images using wavelet based dominant gray level run length texture features. Middle-East J. Sci. Res., 13(7):883-888.
Revathy, N. and Amalraj, R. (2011). Accurate cancer classification using expressions of very few genes. Int. J. Comp. App., 14(4):4.19.
Runtao, D. and Venetsanopoulos, A. (2003). Generalized homomorphic and adaptive order statistic filters for the removal of impulsive and Signal dependent noise. IEEE Trans Circuits Syst., 34:7,948.
Saad, A.E. and El-Samie, A. (2008). Curvelet Fusion of MRI and CT Images. Prog. Electro. Res., 3:9,215.
Sawant, A., Zeman, H., Muratone, D., Samant, S., Di Bianca, F. (1999). Adaptive median filter algorithm to remove impulse noise in x-ray and CT images and speckle in ultrasound images. In: SPIE Proceeding of Medical Imaging: an Diego, CA, United States: p. 11, 1,263.
Stanley, K.O. and Miikkulainen, R. (2002). Evolving neural networks through augmenting topologies. Evolu. Comp., 2(10):28, 99.
Sy Ahmad Ubaidillah, S.H., Sallehuddin, R., and Ali, N.A. (2013). Cancer detection using Artificial Neural Network and Support Vector Machine: A Comparative Study. J. Teknol. (Sci. Eng.), 65(1):73-81.
Wang, L., Lu, J., Li, Y., Yahagi, T., and Okamoto, T. (2005). Noise reduction using wavelet with application to medical X-ray image. International Conference on Industrial Technology, December, 2005: Hong Cong, China: p. 8, 33.
Wang, Y. and Zhou, H. (2006). Total variation wavelet based medical image denoising. Int. J. Biomed. Imag., p. 5, 1.
Wuyang, S., Yaohua, Yi., Ronggang, H., Yong, X. (2019). Robust contrast enhancement forensics based on convolutional neural Networks. Signal Processing: Image Communication, 71:138-146.
Xie, J., Xu, L., and Chen, E. (2012). Image denoising and in painting with deep neural networks. Adv. Neural Inform. Proc. Syst., p. 341-349.
Zhang, K., Zuo, W., Chen, Y., Meng, D., and Zhang, L. (2016). Beyond a Gaussian denoiser: Residual learning of deep CNN for image denoising. arXiv: 1608.03981.
