A robot that can perform guided colonoscopy is presented in a paper published this week in Nature Machine Intelligence. The technology described in this study simplifies automated colonoscopy procedures and can also be applied to other procedures such as pancreatic endoscopy, bronchoscopy and gastroscopy.
Nineteen million colonoscopies are performed every year in the European Union and the United States, and the need for this procedure is expected to rise by 16% in the next decade. The diagnosis and prevention of colorectal cancer — the third most common form of cancer worldwide — relies heavily on colonoscopies. Traditional colonoscopies are expensive, painful and require highly skilled medical practitioners to be fully aware of the probe’s position while guiding it through the colon. New approaches using magnetic probes that are guided by a magnet outside of the body have been recently developed. The external magnet can be precisely controlled by a robotic arm, which allows for complex movement inside the body. However, this method is complicated and still requires the user to be highly trained in order guide the robot arm and the magnetic probe.
James Martin, Bruno Scaglioni and colleagues present a semi-automated robotic system that relies on simple movement commands from the user. The probe moves using machine intelligence and image analysis to automatically guide itself along the colon. The authors tested this approach in an artificial colon model as well as in two pigs. The users performing the procedures found this new system to be much more manageable than the non-intelligent method.
By making colonoscopy easier to perform, this technology might have the potential to increase the ubiquity of the procedure and impact the early detection of colorectal cancer.
Planetary science: Modelling electrolyte transport in water-rich exoplanetsNature Communications
Robotics: Taking millimetre-scale origami robots for a spinNature Communications