Technical Challenges of Real-Time Adaptive MR-Guided Radiotherapy

Abstract

In the past few years, radiotherapy (RT) has experienced a major technological innovation with the development of hybrid machines combining magnetic resonance (MR) imaging and linear accelerators. This new technology for MR-guided cancer treatment has the potential to revolutionize the field of adaptive RT due to the opportunity to provide high-resolution, real-time MR imaging before and during treatment application. However, from a technical point of view, several challenges remain which need to be tackled to ensure safe and robust real-time adaptive MR-guided RT delivery. In this manuscript, several technical challenges to MR-guided RT are discussed. Starting with magnetic field strength tradeoffs, the potential and limitations for purely MR-based RT workflows are discussed. Furthermore, the current status of real-time 3D MR imaging and its potential for real-time RT are summarized. Finally, the potential of quantitative MR imaging for future biological RT adaptation is highlighted.

Magnetic resonance imaging (MRI) guided proton therapy: A review of the clinical challenges, potential benefits and pathway to implementation

“…Gating experience from MRI-Linac systems has shown realtime tracking and gating capabilities using MRI is reliable and accurate with respect to spatial integrity and tracking accuracy [79]. Latency time between image acquisition and gating response needs to be short and has been shown to be between 300 -500 ms on MRI-Linac systems [80]. Cine 2D MRI on MRI-Linac systems has been shown to be sufficient to enable gating, however work towards real-time 3D MRI are in progress.…”

The Potential of Photoacoustic Imaging in Radiation Oncology

“…Nonetheless, these methods have limited spatio-temporal resolution, comparatively long acquisition times, and may require exogenous contrast agents, with associated toxicity ( 35 , 36 ). Furthermore, they have limited capability for deployment in conventional linear accelerator rooms, except through combined MR—linear accelerator systems ( 37 , 38 ) or novel PET—linear accelerator systems ( 39 , 40 ), which are currently limited by high cost and complexity ( 41 ).…”

Accelerating Recurrent Neural Networks for Gravitational Wave Experiments

“…• A novel technique for balancing IIs of multi-layer LSTM inference to increase hardware efficiency and system throughput for data analysis in gravitational wave experiments. • A scalable and low latency LSTM template which enables the generation of low-latency FPGA designs with efficient The specific RNN layered structure and coefficients are LIGO specific, but the need for low latency would benefit many other applications, especially those requiring real-time response, e.g., low latency would benefit the Large Hadron Collider (LHC) physics [14], adaptive radiotherapy [15] and electronic trading [16]. The proposed techniques can be adapted to address these other applications.…”