Deep Learning based Classification for Head and Neck Cancer Detection with Hyperspectral Imaging in an Animal Model

Ling Ma, Guolan Lu, Dongsheng Wang, Xu Wang, Zhuo Georgia Chen, Susan Muller, Amy Chen, Baowei Fei, Ling Ma, Guolan Lu, Dongsheng Wang, Xu Wang, Zhuo Georgia Chen, Susan Muller, Amy Chen, Baowei Fei

Abstract

Hyperspectral imaging (HSI) is an emerging imaging modality that can provide a noninvasive tool for cancer detection and image-guided surgery. HSI acquires high-resolution images at hundreds of spectral bands, providing big data to differentiating different types of tissue. We proposed a deep learning based method for the detection of head and neck cancer with hyperspectral images. Since the deep learning algorithm can learn the feature hierarchically, the learned features are more discriminative and concise than the handcrafted features. In this study, we adopt convolutional neural networks (CNN) to learn the deep feature of pixels for classifying each pixel into tumor or normal tissue. We evaluated our proposed classification method on the dataset containing hyperspectral images from 12 tumor-bearing mice. Experimental results show that our method achieved an average accuracy of 91.36%. The preliminary study demonstrated that our deep learning method can be applied to hyperspectral images for detecting head and neck tumors in animal models.

Keywords: Hyperspectral imaging; convolutional neural networks (CNN); head and neck cancer; machine learning; noninvasive cancer detection; spectral-spatial classification.

Figures

Figure 1.
Figure 1.
Overview of the proposed deep learning method for cancer detection with hyperspectral imaging.
Figure 2.
Figure 2.
The configuration of the convolutional neural networks (CNN).
Figure 3.
Figure 3.
The accuracy in the leave-one-out experiment with different learning rates (from 0.1 to 1), a fixed batch size of 50, and a fixed epoch. Twelve mouse data (#1–12) was used in this experiment.
Figure 4.
Figure 4.
The accuracy in the leave-one-out experiment with different batch sizes (from 10 to 100), a fixed learning rate of 1, and a fixed epoch. Twelve mouse data (#1–12) was used in this experiment.
Figure 5.
Figure 5.
The accuracy in the leave-one-out experiment with different epochs (from 1 to 10), a fixed learning rate of 1, and a fixed batch size of 50. Twelve mouse data (#1–12) was used in this experiment.
Figure 6.
Figure 6.
Qualitative evaluation for the detection of head and neck cancer in three mice. (a) RGB composite images generated from the tumor hypercube, (b) The detection results by the CNN based classification, (c) The final results refined by post-processing, and (d) the pair overlap between our detected result and the gold standard from the GFP images, where the white, black, pink, and green regions mean the true positive, true negative, false positive, and false negative regions, respectively.

References

    1. Mehanna H, et al., “Prevalence of human papillomavirus in oropharyngeal and nonoropharyngeal head and neck cancer—systematic review and meta analysis of trends by time and region,” Head & neck, 35(5), 747–755 (2013).
    1. Haddad RI and Shin DM, “Recent advances in head and neck cancer,” New England Journal of Medicine, 359(11), 1143–1154 (2008).
    1. Liu Z, Wang H, and Li Q, “Tongue tumor detection in medical hyperspectral images,” Sensors, 12(1), 162–174 (2011).
    1. Roblyer D, et al., “Multispectral optical imaging device for in vivo detection of oral neoplasia,” Journal of biomedical optics, 13(2), 024019–024019 (2008).
    1. Roblyer D, et al. “In vivo fluorescence hyperspectral imaging of oral neoplasia,” in SPIE BiOS, Biomedical Optics International Society for Optics and Photonics, 71690J–71690J (2009).
    1. Lu G, et al. “Spectral-spatial classification using tensor modeling for cancer detection with hyperspectral imaging,” in SPIE Medical Imaging, International Society for Optics and Photonics, 903413–903413 (2014).
    1. Lu G, et al. “Spectral-spatial classification for noninvasive cancer detection using hyperspectral imaging,” Journal of biomedical optics, 19(10), 106004–106004 (2014).
    1. Chung H, et al. “Superpixel-based spectral classification for the detection of head and neck cancer with hyperspectral imaging,” in SPIE Medical Imaging, International Society for Optics and Photonics, 978813–978813 (2016).
    1. Krizhevsky A, Sutskever I, and Hinton GE. “Imagenet classification with deep convolutional neural networks,” in Advances in neural information processing systems, 1097–1105 (2012).
    1. Fei B, et al., “MR/PET quantification tools: Registration, segmentation, classification, and MR-based attenuation correction,” Medical physics, 39(10), 6443–6454 (2012).
    1. Akbari H, Fei B “3D ultrasound image segmentation using wavelet support vector machines,” Medical physics, 39(6), 2972–2984 (2012).
    1. Yang X, Wu S, Sechopoulos I, and Fei B “Cupping artifact correction and automated classification for high-resolution dedicated breast CT images,” Medical physics, 39(10), 6397–6406 (2012).
    1. Wang H, and Fei B “An MR image-guided, voxel-based partial volume correction method for PET images,” Medical physics, 39(1), 179–194 (2012).

Source: PubMed

スポンサーと協力者

医学的状態

薬物療法

3
購読する