Deep Recurrent Neural Network Performing Spectral Recurrence on Hyperspectral Images for Brain Tissue Classification
Authors: 
Pedro L. Cebrián, Alberto Martín-Pérez, Manuel Villa, Jaime Sancho, Gonzalo Rosa, Guillermo Vazquez, Pallab Sutradhar, Alejandro Martinez de Ternero, Miguel Chavarrías, Alfonso Lagares, Eduardo Juarez, César SanzAuthors Info & Claims
Pedro L. Cebrián, Alberto Martín-Pérez, Manuel Villa, Jaime Sancho, Gonzalo Rosa, Guillermo Vazquez, Pallab Sutradhar, Alejandro Martinez de Ternero, Miguel Chavarrías, Alfonso Lagares, Eduardo Juarez, César SanzAuthors Info & ClaimsPages 15 - 27
Abstract
Hyperspectral imaging approaches have proven its effectiveness in the medical field for characterizing brain tissues. Furthermore, these techniques in conjunction with machine learning (ML) algorithms has shown to be a useful tool for tumor detection in order to assist neurosurgeons in operations. In this report, it is proposed a novel deep recurrent neural network (DRNN) performing spectral recurrence with the hyperspectral bands to increase precision in tissue classification. In addition, this research present a comparison between the optimized models of the followings ML algorithms: support vector machine, random forest (RF) and the DRNN. All of them were trained by following an hyperparameter optimization process. As a result, DRNN improve brain tissue predictions in terms of the area under the receiver operating characteristic objective test metric (by 1.39% over SVM and 1.91% over RF) whereas RF classification maps illustrate truthfully the distribution of different tissue regions.
References
[1]
Kraus, G.E., et al.: A technique utilizing positron emission tomography and magnetic resonance/computed tomography image fusion to aid in surgical navigation and tumor volume determination. J. Image Guid. Surg., vol. 1, pp. 300–307 (1995).
[2]
ElMasry, G., Sun, D.W.: Principles of hyperspectral imaging technology. In: Hyperspectral Imaging for Food Quality Analysis and Control. Academic Press, p. 3–43 (2010).
[3]
Urbanos G et al. Supervised machine learning methods and hyperspectral imaging techniques jointly applied for brain cancer classification Sensors 2021 21 11 3827
[4]
Zhou Y et al. Crimi A, Bakas S, Kuijf H, Keyvan F, Reyes M, van Walsum T, et al. Holistic brain tumor screening and classification based on DenseNet and recurrent neural network Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries 2019 Cham Springer 208-217
[5]
Mou, L., Ghamisi, P., Zhu, X.X.: deep recurrent neural networks for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 55(7), 3639–3655 .
[6]
Martín-Pérez, A., et al.: Hyperparameter optimization for brain tumor classification with hyperspectral images. In: 2022 25th Euromicro Conference on Digital System Design (DSD) (2022).
[7]
Bengs M, et al., et al. Martel AL, et al., et al. Spectral-spatial recurrent-convolutional networks for In-Vivo hyperspectral tumor type classification Medical Image Computing and Computer Assisted Intervention – MICCAI 2020 2020 Cham Springer 690-699
[8]
Ayaz H et al. Hyperspectral imaging for minced meat classification using nonlinear deep features Appl. Sci. 2020 10 7783
[9]
Knospe L et al. New intraoperative imaging tools and image-guided surgery in gastric cancer surgery Diagnostics 2022 12 507
[10]
Wang, L.: Support Vector Machines: Theory and Applications. Springer Science & Business Media, Auckland, vol. 177 (2005)
[11]
Breiman L Random forests Mach. Learn. 2001 45 5-32
[12]
Svozil D, Kvasnicka V, and Pospichal J Introduction to multi-layer feed-forward neural networks Chemometri. Intell. Lab. Syst. 1997 39 1 43-62
[13]
Pascanu, R., et al.: On the difficulty of training recurrent neural networks. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research, vol. 28, no. 3, pp. 1310–1318 (2013). https://proceedings.mlr.press/v28/pascanu13.html
[14]
Bergstra, J., et al.: Algorithms for hyperparameter optimization. In: Advances in Neural Information Processing Systems, vol. 24 (2011). https://proceedings.neurips.cc/paper_files/paper/2011/file/86e8f7ab32cfd12577bc2619bc635690-Paper.pdf
[15]
Xu Y et al. Hyperspectral image classification via a random patches network ISPRS J. Photogrammetry Remote Sens. 2018 142 344-357
Recommendations
Spectral-spatial classification for hyperspectral image based on a single GRU
AbstractDeep learning methods have been successfully used to extract deep features of many hyperspectral tasks. Multiple neural networks have been introduced in the classification of hyperspectral images, such as convolutional neural network (...
Integrating Convolutional Neural Network and Gated Recurrent Unit for Hyperspectral Image Spectral-Spatial Classification
Pattern Recognition and Computer VisionAbstractIn this paper, we propose a novel deep learning framework for hyperspectral image (HSI) spectral-spatial classification. This framework mainly consists of two components: convolutional neural network (CNN) and gated recurrent unit (GRU). CNN is ...
Hyperspectral Image Features Classification Using Deep Learning Recurrent Neural Networks
The implementation of Deep learning (DL) techniques, Object detection and classification has achieved remarkable results in remote sensing application. Deep learning with Recurrent Neural Network (RNN) technique on hyper-spectral data has been presented ...
Comments
Information & Contributors
Information
Published In
Jan 2023
127 pages
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 27 April 2023
Author Tags
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 06 Oct 2024
Other Metrics
Citations
View Options
View options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in