DblurDoseNet: A deep residual learning network for voxel radionuclide dosimetry compensating for single-photon emission computerized tomography imaging resolution

Keyphrases

• Single-photon Emission Computed Tomography (SPECT) 100%
• Dose Rate 100%
• Computerized Tomography 100%
• Imaging Resolution 100%
• Radionuclide Dosimetry 100%
• Deep Residual Learning 100%
• Residual Learning Network 100%
• Convolutional Neural Network 80%
• Rate Map 80%
• Dose Voxel Kernel 70%
• Phantom 50%
• Monte Carlo 50%
• Virtual Patient 30%
• SPECT-CT 20%
• SPECT Image 20%
• Dosimetry 20%
• Mean Dose 20%
• Normalized Root Mean Square Error 20%
• Residual Learning 20%
• Residual Convolutional Neural Network 20%
• Rate Error 20%
• Clinical Treatment 10%
• Histogram 10%
• Treatment Planning 10%
• Kidney 10%
• Computed Tomography 10%
• CT Images 10%
• Patient-specific 10%
• CT Density 10%
• Tesla 10%
• Spatial Resolution 10%
• Lutetium-177 (177Lu) 10%
• Patient Photographs 10%
• Radionuclide Therapy 10%
• Density Map 10%
• Monte Carlo Radiation Transport 10%
• Monte Carlo Method 10%
• Computational Efficiency 10%
• Activity Diagram 10%
• Tissue Heterogeneity 10%
• Image Map 10%
• Poisson 10%
• U-Net 10%
• Deep Convolutional Neural Network (deep CNN) 10%
• Potential Improvement 10%
• Computationally Expensive 10%
• Neural Network Method 10%
• 177Lu-DOTATATE 10%
• Line Profiles 10%
• Kernel Convolution 10%
• V100 10%
• Patient-specific Dosimetry 10%
• 3D Depth 10%
• Feature Extractor 10%
• Resolution Effect 10%
• Noise Standard Deviation 10%
• Density Scaling 10%
• Single GPU 10%
• Depth Feature 10%

Medicine and Dentistry

• Computer Assisted Tomography 100%
• Radioisotope 100%
• Dosimetry 100%
• Single-Photon Emission Computed Tomography 75%
• Virtual Patient 75%
• Single Photon Emission Computed Tomography-Computed Tomography 50%
• Lutetium 177 50%
• Treatment Planning 25%
• Radionuclide Therapy 25%