Wavelength Bands Reduction Method in Near-Infrared Hyperspectral Image based on Deep Neural Network for Tumor Lesion Classification
Keywords:Near-infrared, Hyperspectral imaging, Machine learning, Wavelength reduction
In this paper, we propose a method for wavelength bands reduction of near-infrared (NIR) hyperspectral imaging data to extract the cancer region with minimum input data. NIR hyperspectral imaging data has a spectrum data of each pixel and is suitable for distinguishing tumors region of the body rather than RGB imaging data. However, it is difficult to applicate to the medical field because of processing time consumption and hardware size limitation. Therefore, it is necessary to remove the redundant wavelength bands which are not (or little) contributed to tumor region extraction. Although several previous studies for wavelength bands reduction have been conducted, these approaches focused on the characteristics of the wavelength itself. In this research, the proposed wavelength bands reduction method is focused on the node weights of the post-training deep neural network which is an indicator directly related to classification. The experimental results using GIST specimen demonstrated that four wavelength bands selected from all wavelengths bands by using the proposed method are effective for tumor distinguishing as well as all wavelength bands.
(1) Schultz R. A., Nielsen T., Zavaleta J. R., Ruch R., Wyatt R. and Ganner H. R., Hyperspectral Imaging: A Novel Approach For Microscopic Analyasis, Cytometry, 2001. 43(4): p. 239-247.
(2) Chein-I Chang, Hyperspetral imaging: teqchniques for apectral detection and classification, Springer, 2003
(3) Douglas Barbin, Gamal Elmasry, Da-Wen Sun and Paul Allen, Near-infrared hyperspectral imaging for grading and classification of pork, Meat Science, 2012. 90(1): p. 259-268
(4) Svetlana V. Panasyuk, Shi Yang, Douglas V. Faller, Duyen Ngo, Robert A. Lew, Jenny E. Freeman and Adrianne E. Rogers, Medical hyperspectral imaging to facilitate residual tumor identification during surgery, Cancer Biology & Therapy, 2007. 6(3): p. 439-446
(5) Kho, E., de Boer, L.L., Van de Vijver, K.K., van Duijnhoven, F., Vrancken Peeters, M.T.F.D., Sterenborg, H.J.C.M. et al. Hyperspectral imaging for resection margin assessment during cancer surgery. Clin Cancer Res. 2019, 25: p. 3572-3580
(6) Tuan Vo-Dinh, et al., A hyperspectral imaging system for in vivo optical diagnostics, IEEE Engineering in Medicine and Biology Magazine, 2004. 23(5) p. 40-49
(7) D. Sato, T. Takamatsu, M. Umezawa, Y. Kitagawa, K. Maeda, N. Hosokawa, K. Okubo, M. Kamimura, T. Kadota, T. Akimoto, T. Kinoshita, T. Yano, T. Kuwata, H. Ikematsu, H. Takemura, H. Yokota and K. Soga, Distinction of surgically resected gastrointestinal stromal tumor by near-infrared hyperspectral imaging, Scientific Reports, (in press)
(8) Zako T., Ito M., Hyodo H., Yoshimoto M., Watanabe M., Takemura H., Kishimoto H., Kaneko K., Soga K. and Maeda M., Extra-iluminal detection of assumed colonic tumor site by near-infrared laparoscopy, Surgical Endoscopy, 2015, p. 1-7.
(9) Zako T., Hyodo H., Tsuji K., Tokuzen K., Kishimoto H., Ito M., Kaneko K., Maeda M. and Soga K., Development of near infrared-fluorescent nanophosphors and applications for cancer diagnosis and therapy, Journal of Nanomaterials, 2010. 2010, p.1-7.
(10) Elisabeth J. M. Baltussen, Esther N. D. Kok, Susan G. Brouwer de Koning, Joyce Sanders, Arend G.J. Aalbers, Niels F. M. Kok, Geerard L. Beets, Claudie C. Flohil, Sjoerd C. Bruin, Koert F. D. Kuhlmann, Henricus J. C. M. Sterenborg, Theo J. M. Ruers, Hyperspectral imaging for tissue classification, a way toward smart laparoscopic colorectal surgery, J. Biomed. Opt. 2019, 24(1): 016002
(11) Hamed Akbari, Kuniaki Uto, Yukio Kosugi, Kazuyuki Kojima and Naofumi Tanaka, Cancer detection using infrared hyperspectral imaging, Cancer Science, 2011. 102(4): p. 852-857
(12) Guolan Lu and Baowei Fei, Medical hyperspectral imaging: a review, Journal of Biomedical Optics, 2014. 19(1): 010901.
(13) Jianchang Ren, Jaime Zabalza, Stephen Marshall, Jiangbin Zheng, Effective Feature Extraction and Data Reduction in Remote Sensing Using Hyperspectral Imaging, IEEE Signal Processing Magazine, 2014, 31(4): p. 149-154
(14) M. Kasai, Y. Yasuda, H. Takemura, H. Mizoguchi, K. Soga, K. Kaneko M. Kasai, In Vivo Tumor Spatial Classification using PCA and K-Means with NIR-Hyperspectral Data, J. Biomed. Eng. and Med. Imaging, 2016, 3, 2055.
(15) Silvia Serranti, Daniela Cesare, Federico Marini and Giuseppe Bonifazi, Classification of oat and groat kernels using NIR hyperspectral imaging, Talenta, 2013. 103: p. 276-284
(16) Daniel Smilkov, Nikhil Thorat, Been Kim, Fernanda Viegas and Martin Wattenberg, SmoothGrad: removing noise by adding noise, arXiv:1706.03825, 2017
(17) Ramprasaath R. Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh and Dhruv Batra, Grad-CAM: Visual Explanations From Deep Networks via Gradient-Based Localization, Proceedings of the IEEE International Conference on Computer Vision, 2017: p.618-626
(18) Jost Tobias Springenberg, Alexey Dosovitskiy, Thomas Brox and Martin Riedmiller, Striving for Simplicity: The All Convolutional Net, arXiv:1412.6806, 2014
(19) David E. Rumelhart. Richard Durbin, Richard Golden and Yves Chauvin, Backpropagation: The Basic Theory, Psychology Press, 1995