3D Convolutional Neural Network for Voxel-level Prediction of Radiation Absorption (CMIMI 2018 Presentation)
Myeongchan Kim (MD) gave a presentation on his paper “3D Convolutional Neural Network for Voxel-level Prediction of Radiation Absorption” at 2018 SIIM Conference on Machine Intelligence in Medical Imaging (CMIMI).
Kim, M., Li, X., Yune, S., Lee, H., Liu, B., Do, S., 2018. 3D Convolutional Neural Network for Voxel-level Prediction of Radiation Absorption.
The risk of diagnostic imaging procedures such as computed tomography (CT) can be best assessed by measuring the radiation dose absorbed by individual organs. For organ-specific dose estimation, Monte Carlo simulation of radiation transport (MCRT) have been widely used without having to take in-vivo dose measurements. However, MCRT requires substantial computing power and simulation time, limiting its application in daily clinical practice. As a method to build robust prediction models without rigorous computation for every case, machine-learning techniques using convolutional neural networks (CNN) have been extensively explored and have succeeded in many fields recently. In this study, we demonstrate a novel method for voxel-level prediction of radiation dose by using a 3D CNN.