DeepDRR -- A Catalyst for Machine Learning in Fluoroscopy-guided Procedures

TL;DR

DeepDRR is a fast, realistic simulation framework that generates synthetic fluoroscopy images from CT scans, enabling effective machine learning training without extensive real data annotation.

Contribution

We introduce DeepDRR, a novel framework combining machine learning and analytic methods to simulate fluoroscopy images from CT scans for training purposes.

Findings

Models trained on DeepDRRs generalize to real clinical data

DeepDRR enables training without re-training or domain adaptation

Promising results support machine learning in fluoroscopy-guided procedures

Abstract

Machine learning-based approaches outperform competing methods in most disciplines relevant to diagnostic radiology. Interventional radiology, however, has not yet benefited substantially from the advent of deep learning, in particular because of two reasons: 1) Most images acquired during the procedure are never archived and are thus not available for learning, and 2) even if they were available, annotations would be a severe challenge due to the vast amounts of data. When considering fluoroscopy-guided procedures, an interesting alternative to true interventional fluoroscopy is in silico simulation of the procedure from 3D diagnostic CT. In this case, labeling is comparably easy and potentially readily available, yet, the appropriateness of resulting synthetic data is dependent on the forward model. In this work, we propose DeepDRR, a framework for fast and realistic simulation of fluoroscopy and digital radiography from CT scans, tightly integrated with the software platforms native to deep learning. We use machine learning for material decomposition and scatter estimation in 3D and 2D, respectively, combined with analytic forward projection and noise injection to achieve the required performance. On the example of anatomical landmark detection in X-ray images of the pelvis, we demonstrate that machine learning models trained on DeepDRRs generalize to unseen clinically acquired data without the need for re-training or domain adaptation. Our results are promising and promote the establishment of machine learning in fluoroscopy-guided procedures.