Boosting-based Cascaded Convolutional Neural Networks for the Segmentation of CT Organs-at-risk in Nasopharyngeal Carcinoma

Tao Zhong; Xia Huang; Fan Tang; Shujun Liang; Xiaogang Deng; Yu Zhang

doi:10.1002/mp.13825

Boosting-based Cascaded Convolutional Neural Networks for the Segmentation of CT Organs-at-risk in Nasopharyngeal Carcinoma

Med Phys. 2019 Sep 16. doi: 10.1002/mp.13825. Online ahead of print.

Authors

Tao Zhong^{1

2}, Xia Huang^{1

3}, Fan Tang^{1

2

4}, Shujun Liang^{1

2}, Xiaogang Deng⁴, Yu Zhang^{1

2}

Affiliations

¹ School of Biomedical Engineering, Southern Medical University, 1838 North Guangzhou Avenue, Guangzhou, 510515, Guangdong, China.
² Guangdong Provincial Key Laboratory of Medical Image Processing, Southern Medical University, 1838 North Guangzhou Avenue, Guangzhou, 510515, Guangdong, China.
³ Department of Medical Imaging Center, Nanfang Hospital, Southern Medical University, 1838 North Guangzhou Avenue, Guangzhou, 510515, Guangdong, China.
⁴ Department of Radiation Oncology, Nanfang Hospital, Southern Medical University, 1838 North Guangzhou Avenue, Guangzhou, 510515, Guangdong, China.

PMID: 31529501
DOI: 10.1002/mp.13825

Abstract

Purpose: Accurately segmenting organs-at-risk (OARs) is a key step in the effective planning of radiation therapy for nasopharyngeal carcinoma (NPC) treatment. In OAR segmentation of the head and neck CT, the low contrast and surrounding adhesion tissues of the parotids, thyroids, and optic nerves result in the difficulty in segmentation and lower accuracy of automatic segmentation for these organs than the other organs. In this paper, we propose a cascaded network structure to delineate these three OARs for NPC radiotherapy by combining deep learning and Boosting algorithm.

Materials and methods: The CT images of 140 NPC patients treated with radiotherapy were collected, and each of the three OAR annotations was respectively delineated by an experienced rater and reviewed by a professional radiologist (with 10 years of experience). The datasets (140 patients) were divided into a training set (100 patients), a validation set (20 patients), and a test set (20 patients). From the Boosting method for combining multiple classifiers, three cascaded CNNs for segmentation were combined. The first network was trained with the traditional approach. The second one was trained on patterns (pixels) filtered by the first net. That is, the second machine recognized a mix of patterns (pixels), 50% of which was accurately identified by the first net. Finally, the third net was trained on the new patterns (pixels) screened jointly by the first and second networks. During the test, the outputs of the three nets were considered to obtain the final output. Dice similarity coefficient (DSC), 95th percentile of the Hausdorff distance (95% HD), and volume overlap error (VOE) were used to assess the method performance.

Results: The mean DSC (%) values were above 92.26 for the parotids, above 92.29 for the thyroids, and above 89.37 for the optic nerves. The mean 95% HDs (mm) were approximately 3.08 for the parotids, 2.64 for the thyroids, and 2.03 for the optic nerves. The mean VOE (%) values were approximately 14.16 for the parotids, 14.94 for the thyroids, and 19.07 for the optic nerves.

Conclusion: The proposed cascaded deep learning structure could achieve high performance compared with existing single-network or other segmentation algorithms.

Keywords: Boosting; Deep learning; Nasopharyngeal carcinoma; Organs-at-risk; Segmentation.