An autonomous lung robot could revolutionise the way doctors approach lung cancer treatment.
Scientists have shown that a steerable lung robot can manoeuvre the intricacies of the lung and avoid critical structures in order to deliver treatment.
Lung cancer poses a significant challenge for clinicians due to the presence of small, hidden tumours deep within the tissue.
Researchers from UNC-Chapel Hill and Vanderbilt University in the US have developed a flexible, robust robot for traversing lung tissue.
The team’s research, published in Science Robotics, showcases the capabilities of the lung robot.
Designed by Dr Ron Alterovitz, of the UNC Department of Computer Science, and Dr Jason Akulian, of the UNC Department of Medicine, the autonomous medical needle navigates from ‘Point A’ to ‘Point B’ within a living laboratory model while meticulously avoiding vital structures such as tiny airways and blood vessels.
A mechanical control system provides precise thrust for the needle’s movement, allowing it to navigate forward and backward.
The needle, made from a nickel-titanium alloy, is laser-etched to enhance its flexibility, enabling it to glide effortlessly through lung tissue.
As the needle advances, its etched design facilitates smooth navigation around obstacles. Additional attachments, such as catheters, can be used in conjunction with the needle to perform procedures such as lung biopsies.
To navigate through lung tissue, the robot relies on a combination of CT scans of the patient’s thoracic cavity and AI. These tools create 3D models of the lung, including airways, blood vessels and the target area.
Once the needle is positioned for launch, the AI-driven software directs it to autonomously travel from Point A to Point B whilst skilfully avoiding crucial structures.
Dr Akulian, co-author of the paper and Section Chief of Interventional Pulmonology and Pulmonary Oncology, said: ‘This technology allows us to reach targets we can’t otherwise reach with a standard or even robotic bronchoscope. It gives you that extra few centimetres or few millimetres, which help immensely with pursuing small targets in the lungs.’
The development is thanks to collaboration at UNC, bringing together medical, computer science and engineering experts.
Alongside Drs Alterovitz and Akulian, the team included Dr Yueh Z Lee from the UNC Department of Radiology, Robert J Webster III from Vanderbilt University and Alan Kuntz from the University of Utah.
Dr Alterovitz, the principal investigator on the project, likened the robot’s capabilities to a self-driving car – navigating through lung tissue and adeptly avoiding obstacles like major blood vessels as it reaches its destination.
Because the lungs continually expand and contract within the chest cavity, it presents an additional challenge as the robot must track moving targets. Dr Akulian compared it to ‘shooting at a moving target’.
To address this challenge, the researchers tested their robot in conjunction with a laboratory model that simulated intermittent breath-holding.
The robot was programmed to advance during these periods, adapting to the model’s respiratory motion.
While the team acknowledges that there are still nuances to refine regarding target acquisition and effectiveness, they say the range of possibilities is promising.
Photo caption: Overview of the semi-autonomous medical robot’s three stages in the lungs.
Stage 1 (top) consists of a physician manually navigating a traditional bronchoscope through the airways.
Stage 2 (middle) consists of a physician teleoperating the aiming device.
Stage 3 (bottom) consists of autonomous needle deployment through the parenchyma to the nodule while accounting for respiratory motion and avoiding anatomical obstacles such as clinically relevant vasculature, bronchi and the visceral pleural boundary.
