Room: Room 207
Purpose: Deep learning excels conventional methods in the domain of image recognition, classification and segmentation. Instead of predicting DVHs with machine learning as is conventionally done, here we propose to predict the isodose in voxel domain with the popular Fully Convolution Networks(FCN).
Methods: We trained the FCN end-to-end, pixels-to-pixels on over 10,000 input images with artificially generated patient contours and corresponding output images with dose distributions, in a resolution of 512x512. We generate the BODY, PTV and OAR contours randomly to increase the variety of the training set, thus testing the robustness of the method. The location, size and shape of the contours are different from case to case. We calculate the dose distribution with a pencil beam algorithm for a 6x beam. We optimized a 3D conformal plan by balancing the mean dose to the PTV and the OAR. The first 5layers of the FCN consist of convolution and pooling layer that generate 1d vector representing the image classes. The last three layers of the FCN are all deconvolutional layers which up-sample the coarse output to the same resolution as the original input. The network was trained by Stochastic Gradient Descent in 40epochs. We used the Caffe framework and matlab for our deep learning implementation.Mean Intersection over union (MIOU) and pixel accuracy (PA) are used to test the accuracy on a validation set that is generated the same way as the training set with 1000 cases.
Results: The MIOU is ~63% and the PA is ~91% on the validation set. The dose prediction takes around 0.1 second.
Conclusion: We showed the potential of deep learning method in reliably predicting dose distribution from patient geometries alone. The versatility and speed of this method makes it a valuable tool for any TPS and promises to significantly improve the clinical workflow.