Segmentation of the prostate and organs at risk in male pelvic CT images using deep learning

Abstract

Inter-and intra-observer variation in delineating regions of interest (ROIs) occurs because of differences in expertise level and preferences of the radiation oncologists. We evaluated the accuracy of a segmentation model using the U-Net structure to delineate the prostate, bladder, and rectum in male pelvic CT images. The dataset used for training and testing the model consisted of raw CT scan images of 85 prostate cancer patients. We designed a 2D U-Net model to directly learn a mapping function that converts a 2D CT grayscale image to its corresponding 2D OAR segmented image. Our network contains blocks of convolution 2D layers with variable kernel sizes, channel number, and activation functions. On the left side of the U-Net model, we used three 3x3 convolutions, each followed by a rectified linear unit (ReLu) (activation function), and one max pooling operation. On the right side of the U-Net model, we used a 2x2 transposed convolution and two 3x3 convolution networks followed by a ReLu activation function. The automatic segmentation using the U-Net generated an average dice similarity coefficient (DC) and standard deviation (SD) of the following: DC ± SD (0.88 ± 0.12), (0.95 ± 0.04), and (0.92 ± 0.06) for the prostate, bladder, and rectum, respectively. Furthermore, the mean of average surface Hausdorff distance (ASHD) and SD were 1.2 ± 0.9 mm, 1.08 ± 0.8 mm, and 0.8 ± 0.6 mm for the prostate, bladder, and rectum, respectively. Our proposed method, which employs the U-Net structure, is highly accurate and reproducible for automated ROI segmentation. This provides a foundation to improve automatic delineation of the boundaries between the target and surrounding normal soft tissues on a standard radiation therapy planning CT scan.

CT prostate segmentation based on synthetic MRI‐aided deep attention fully convolution network

“…Kazemifar et al . used a 2D U‐Net structure, a DNN variant, to delineate the prostate, bladder, and rectum in pelvic CT images . However, this method was based on 2D inputs, which lacks 3D spatial information.…”

“…15 Deep features, which are directly assigned with semantic and structural information of the prostate region, are used to train a deep learning-based classifier. Liu 19 However, this method was based on 2D inputs, which lacks 3D spatial information. Balagopal et al presented a fully automated workflow for male pelvic CT image segmentation using deep learning.…”

Clinical evaluation of a full-image deep segmentation algorithm for the male pelvis on cone-beam CT and CT

“…This allows to increase the number of filters being used and the semantic segmentation of many different problems has become feasible using very similar algorithms. For example, similar deep neural networks have been used for segmenting organs in the pelvic [4], abdominal [5], thoracical [6] and head and neck [7] region as well as for medical object detection [8].…”

A convolutional neural network algorithm for automatic segmentation of head and neck organs at risk using deep lifelong learning

“…Deep convolutional neural networks (CNNs) have recently been shown to be well‐suited for image classification problems based on unstructured highly dimensional data . Recently, axial two‐dimensional (2D) CNN strategies have shown promise for autosegmentation in radiotherapy . However, 2D convolutions disregard information about neighboring slices, which can lead to prediction errors in less well‐defined organs.…”

“…[14][15][16][17][18] Recently, axial two-dimensional (2D) CNN strategies have shown promise for autosegmentation in radiotherapy. [19][20][21][22][23][24] However, 2D convolutions disregard information about neighboring slices, which can lead to prediction errors in less well-defined organs. In response, some groups have implemented orthogonal 2D CNNs into their prediction strategies.…”