101. Deep learning-based three-dimensional segmentation of the prostate on computed tomography images
- Author
-
Martin Halicek, Baowei Fei, David M. Schuster, James D. Dormer, and Maysam Shahedi
- Subjects
medicine.diagnostic_test ,business.industry ,Image-Guided Procedures, Robotic Interventions, and Modeling ,Deep learning ,Pattern recognition ,Computed tomography ,Image segmentation ,3D modeling ,Convolutional neural network ,medicine.anatomical_structure ,Similarity (network science) ,Prostate ,Medicine ,Radiology, Nuclear Medicine and imaging ,Segmentation ,Artificial intelligence ,business - 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 [Formula: see text] for Dice similarity coefficient (DSC), [Formula: see text] for mean absolute distance (MAD), and [Formula: see text] for signed volume difference ([Formula: see text]). The average recorded interexpert difference measured on the same test dataset was 92% (DSC), 1.1 mm (MAD), and [Formula: see text] ([Formula: see text]). The proposed algorithm is fast, accurate, and robust for 3-D segmentation of the prostate on CT images.
- Published
- 2019