Segmentation of organs-at-risks in head and neck CT images using convolutional neural networks

Abstract

Purpose: Accurate segmentation of organs-at-risks (OARs) is the key step for efficient planning of radiation therapy for head and neck (HaN) cancer treatment. In the work, we proposed the first deep learning-based algorithm, for segmentation of OARs in HaN CT images, and compared its performance against state-of-the-art automated segmentation algorithms, commercial software, and interobserver variability.

Methods: Convolutional neural networks (CNNs)-a concept from the field of deep learning-were used to study consistent intensity patterns of OARs from training CT images and to segment the OAR in a previously unseen test CT image. For CNN training, we extracted a representative number of positive intensity patches around voxels that belong to the OAR of interest in training CT images, and negative intensity patches around voxels that belong to the surrounding structures. These patches then passed through a sequence of CNN layers that captured local image features such as corners, end-points, and edges, and combined them into more complex high-order features that can efficiently describe the OAR. The trained network was applied to classify voxels in a region of interest in the test image where the corresponding OAR is expected to be located. We then smoothed the obtained classification results by using Markov random fields algorithm. We finally extracted the largest connected component of the smoothed voxels classified as the OAR by CNN, performed dilate-erode operations to remove cavities of the component, which resulted in segmentation of the OAR in the test image.

Results: The performance of CNNs was validated on segmentation of spinal cord, mandible, parotid glands, submandibular glands, larynx, pharynx, eye globes, optic nerves, and optic chiasm using 50 CT images. The obtained segmentation results varied from 37.4% Dice coefficient (DSC) for chiasm to 89.5% DSC for mandible. We also analyzed the performance of state-of-the-art algorithms and commercial software reported in the literature, and observed that CNNs demonstrate similar or superior performance on segmentation of spinal cord, mandible, parotid glands, larynx, pharynx, eye globes, and optic nerves, but inferior performance on segmentation of submandibular glands and optic chiasm.

Conclusion: We concluded that convolution neural networks can accurately segment most of OARs using a representative database of 50 HaN CT images. At the same time, inclusion of additional information, for example, MR images, may be beneficial to some OARs with poorly visible boundaries.

Keywords: convolutional neural networks; deep learning; head and neck; radiotherapy; segmentation.

Figure 1

A schematic illustration of the convolutional neural network architecture. Three orthogonal cross‐sections around target voxel define the input of the network that consists of three stacks of convolution, ReLU, max‐pooling layer, and dropout layers, fully connected and softmax layers.

Figure 2

The strongest gradients of the target CT image (first raw) intersect around the center of the patient's skull (second raw).

Figure 3

A schematic set of parameters and commands used to define convolutional neural network for segmentation of organ‐at‐risks in head‐and‐neck CT images. The parameters in bold correspond to the size of convolution and pooling layers.

Figure 4

Box plot results of convolutional neural network‐based segmentation of organs‐at‐risks in head and neck CT images reported in terms of Dice coefficient.

Figure 5

Segmentation results for three (a–c) selected head and neck CT image, shown in four axial cross‐sections. The reference segmentations are depicted in green and convolution neural network‐based segmentations are depicted in red.