Real-time MRI navigation for magnetic robots: a step forward in minimally invasive therapy
Higher Education Press
image: The integration of robotic drive and imaging feedback is implemented using MRI sequences. The entire cycle time is reduced to 30 ms for real-time navigation, and the introduction of reconstruction algorithms enables the removal of artifacts.
Credit: Renkuan Zhai, Zhangqi Pan, Yuanshi Kou, Chuang Yang, Yang Ruan, Chenli Xu, Linjie He, Jianfeng Zang
A new study published in Engineering presents a novel approach to real-time and artifact-free magnetic resonance imaging (MRI) navigation for magnetic robots, offering significant advancements in the field of minimally invasive medical procedures. Researchers from Huazhong University of Science and Technology in China have developed a multi-frequency dual-echo (MFDE) MRI sequence that enables precise tracking of magnetic robots with a repetition time (TR) of just 30 milliseconds, a substantial improvement over conventional MRI sequences.
Magnetic microrobots hold great promise for minimally invasive therapies, targeted drug delivery, and theranostic interventions due to their small size, flexibility, and targeting capabilities. However, real-time control and high-resolution imaging have long been challenging due to the inherent limitations of MRI sequences, which typically require long repetition times (around 1000 milliseconds) to produce clear images. This delay not only affects the accuracy of robot tracking but also limits the duty cycle of the driving gradient, making real-time manual intervention difficult.
The research team addressed these challenges by introducing a multi-frequency excitation sequence with dual-echo (MFDE) that reduces TR to 30 milliseconds. This sequence employs two adjacent 180° radio-frequency (RF) pulses to generate dual echoes, significantly accelerating the proton spin recovery process. Additionally, alternating positive and negative offset frequency excitations are used to overcome the low signal intensity caused by the steady-state phenomenon, ensuring high-quality imaging even at such a short TR.
The study demonstrates that the MFDE sequence achieves precise positioning of magnetic particles with a relative error of less than 1%, while the duty cycle of the driving gradient reaches 77%. This eliminates the influence of imaging gradients on robot motion, providing artifact-free background imaging. The researchers further developed a reconstruction algorithm that replaces artifacts with bright spots on a pre-obtained background, enabling real-time feedback and precise control of the magnetic robot.
To validate the feasibility and effectiveness of their approach, the researchers conducted experiments in various settings, including maze navigation, phantom endovascular navigation, and in vivo trials in a rat large intestine. In the maze navigation experiment, the magnetic robot was successfully guided through a complex three-dimensional maze, with real-time updates of its position displayed on a three-view imaging platform. The robot's movement was controlled using a joystick, allowing for precise adjustments in real-time.
In the phantom endovascular navigation, the magnetic robot was able to navigate through tortuous vessels with high accuracy, demonstrating the potential for vascular interventions. The in vivo trials in a rat large intestine further showcased the technology's ability to navigate through complex biological environments, providing a promising alternative to traditional colonoscopy procedures.
This study represents a significant step forward in MRI-driven robotic navigation, addressing the critical balance between imaging speed and quality. The development of the MFDE sequence and the accompanying imaging-control platform offer a new avenue for real-time, artifact-free navigation of magnetic robots, paving the way for more precise and efficient minimally invasive medical procedures.
The paper “Multi-Frequency Dual-Echo Magnetic Resonance Imaging for Real-Time and Artifact-Free Magnetic Robot Navigation,” is authored by Renkuan Zhai, Zhangqi Pan, Yuanshi Kou, Chuang Yang, Yang Ruan, Chenli Xu, Linjie He, Jianfeng Zang. Full text of the open access paper: https://doi.org/10.1016/j.eng.2025.04.027. For more information about Engineering, visit the website at https://www.sciencedirect.com/journal/engineering.
Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.