Abstract
Personalized dosimetry with high accuracy is crucial owing to the growing interests in personalized medicine. The direct Monte Carlo simulation is considered as a state-of-art voxel-based dosimetry technique; however, it incurs an excessive computational cost and time. To overcome the limitations of the direct Monte Carlo approach, we propose using a deep convolutional neural network (CNN) for the voxel dose prediction. PET and CT image patches were used as inputs for the CNN with the given ground truth from direct Monte Carlo. The predicted voxel dose rate maps from the CNN were compared with the ground truth and dose rate maps generated voxel S-value (VSV) kernel convolution method, which is one of the common voxel-based dosimetry techniques. The CNN-based dose rate map agreed well with the ground truth with voxel dose rate errors of 2.54% ± 2.09%. The VSV kernel approach showed a voxel error of 9.97% ± 1.79%. In the whole-body dosimetry study, the average organ absorbed dose errors were 1.07%, 9.43%, and 34.22% for the CNN, VSV, and OLINDA/EXM dosimetry software, respectively. The proposed CNN-based dosimetry method showed improvements compared to the conventional dosimetry approaches and showed results comparable with that of the direct Monte Carlo simulation with significantly lower calculation time.
Highlights
Internal dosimetry is becoming increasingly important owing to the growing interest in targeted radionuclide therapies, radiotheranostics, and personalized medicine[1,2,3]
The absolute 3D radioactivity distribution given by quantitative positron emission tomography (PET) or single photon emission tomography (SPECT) and media property derived from transmission scans, such as X-ray computed tomography (CT), are fed to the convolutional neural network (CNN) as an input, and the CNN is trained to generate the dose rate map as an output, with the Monte Carlo simulation based dose rate map as the reference
The whole-body dose rate maps generated by direct Monte Carlo simulation, which is considered as the ground truth, were compared with conventional voxel S-value (VSV) kernel convolution and the proposed approach
Summary
Internal dosimetry is becoming increasingly important owing to the growing interest in targeted radionuclide therapies, radiotheranostics, and personalized medicine[1,2,3]. Voxel-based dosimetry techniques that consider heterogeneous activity distributions have been suggested, including the dose point kernel[7,8,9] and voxel S-value (VSV) approaches[10]. For more accurate personalized dosimetry, voxel-based dosimetry based on direct Monte Carlo simulation that can consider both heterogeneous activities and medium distributions has been suggested. The Monte Carlo simulation generates and tracks particles at the voxel-level and calculates deposited energy to estimate the voxel-level absorbed doses[16,17] This approach requires extensive computational time and resources; it is rarely used in a clinical routine basis. We suggest a new internal radiation dose calculation method, called Deep-dose, which applies a convolutional neural network (CNN) to estimate the voxel dose values from the individual nuclear medicine images. We evaluated the performance of the Deep-dose by comparing its results to those of direct Monte Carlo based voxel dose estimation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
