Real-Time 2D-3D Deformable Registration with Deep Learning and Application to Lung Radiotherapy Targeting

TL;DR

This paper introduces a deep learning method for real-time 2D-3D deformable registration to track lung tumors during radiotherapy, reducing invasiveness and treatment time by avoiding fiducial markers.

Contribution

It develops a patient-specific motion subspace and a deep neural network to recover anatomical positions from a single fluoroscopic image in real-time, enabling accurate lung deformation modeling.

Findings

Geometric accuracy comparable to traditional image registration.

Real-time deformation recovery from single fluoroscopic projection.

Application to lung radiotherapy targeting.

Abstract

Radiation therapy presents a need for dynamic tracking of a target tumor volume. Fiducial markers such as implanted gold seeds have been used to gate radiation delivery but the markers are invasive and gating significantly increases treatment time. Pretreatment acquisition of a respiratory correlated 4DCT allows for determination of accurate motion tracking which is useful in treatment planning. We design a patient-specific motion subspace and a deep convolutional neural network to recover anatomical positions from a single fluoroscopic projection in real-time. We use this deep network to approximate the nonlinear inverse of a diffeomorphic deformation composed with radiographic projection. This network recovers subspace coordinates to define the patient-specific deformation of the lungs from a baseline anatomic position. The geometric accuracy of the subspace deformations on real patient data is similar to accuracy attained by original image registration between individual respiratory-phase image volumes.