Deep learning-based three-dimensional segmentation of the prostate on computed tomography images

Abstract

Segmentation of the prostate in computed tomography (CT) is used for planning and guidance of prostate treatment procedures. However, due to the low soft-tissue contrast of the images, manual delineation of the prostate on CT is a time-consuming task with high interobserver variability. We developed an automatic, three-dimensional (3-D) prostate segmentation algorithm based on a customized U-Net architecture. Our dataset contained 92 3-D abdominal CT scans from 92 patients, of which 69 images were used for training and validation and the remaining for testing the convolutional neural network model. Compared to manual segmentation by an expert radiologist, our method achieved $83\%\pm 6\%$ for Dice similarity coefficient (DSC), $2.3\pm 0.6\mathrm{mm}$ for mean absolute distance (MAD), and $1.9\pm 4.0{\mathrm{cm}}^3$ for signed volume difference ( $\Delta V$ ). The average recorded interexpert difference measured on the same test dataset was 92% (DSC), 1.1 mm (MAD), and $2.1{\mathrm{cm}}^3$ ( $\Delta V$ ). The proposed algorithm is fast, accurate, and robust for 3-D segmentation of the prostate on CT images.

Keywords: computed tomography; convolutional neural network; deep learning; image segmentation; prostate.