Automatic Tumor Segmentation with Deep Convolutional Neural Networks for Radiotherapy Applications

• Published in: Neural processing letters Vol. 48; no. 3; pp. 1323 - 1334
• Main Authors: Wang, Yan, Zu, Chen, Hu, Guangliang, Luo, Yong, Ma, Zongqing, He, Kun, Wu, Xi, Zhou, Jiliu
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.12.2018; Springer Nature B.V
• Subjects: Artificial Intelligence; Artificial neural networks; Brain cancer; Brain research; Classification; Clinical medicine; Complex Systems; Computational Intelligence; Computer Science; Image contrast; Image segmentation; Labels; Machine learning; Magnetic resonance imaging; Medical imaging; Neural networks; Radiation therapy; Support vector machines; Tumors; Southeast Asia; Nasopharyngeal carcinoma (NPC); Magnetic resonance images (MRI); Deep convolutional neural network; 3D image; Medical image segmentation
• ISSN: 1370-4621; 1573-773X
• DOI: 10.1007/s11063-017-9759-3

Accurate tumor delineation in medical images is of great importance in guiding radiotherapy. In nasopharyngeal carcinoma (NPC), due to its high variability, low contrast and discontinuous boundaries in magnetic resonance images (MRI), the margin of the tumor is especially difficult to be identified, making the radiotherapy planning a more challenging problem. The objective of this paper is to develop an automatic segmentation method of NPC in MRI for radiosurgery applications. To this end, we present to segment NPC using a deep convolutional neural network. Specifically, to obtain spatial consistency as well as accurate feature details for segmentation, multiple convolution kernel sizes are employed. The network contains a large number of trainable parameters which capture the relationship between the MRI intensity images and the corresponding label maps. When trained on subjects with pre-labeled MRI, the network can estimate the label class of each voxel for the testing subject which is only given the intensity image. To demonstrate the segmentation performance, we carry on our method on the T1-weighted images of 15 NPC patients, and compare the segmentation results against the radiologist’s reference outline. Experimental results show that the proposed method outperforms the traditional hand-crafted features based segmentation methods. The presented method in this paper could be useful for NPC diagnosis and helpful for guiding radiotherapy.