Brain surgery--robotic medical imaging guides the way
Andy Wilson Editor-at-Large
At the recent International Robots & Vision Show and Conference (Detroit, MI; May 11-13), attendees received a close-up view of the combined efforts of third-party OEM vision companies and robotic-systems vendors (see p. 30). Now, medical-imaging-system vendors may soon offer the same automated capabilities using computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) imagery data to direct sophisticated robot surgeons. In addition to giving physicians advanced operating techniques and tools, virtual-reality (VR) techniques can help reduce the risks and costs of surgical procedures, say researchers.
"VR separates traditional computer interfaces from physical realities, allowing the user to study a model at any scale and orientation without studying the process needed to generate that model," says Dr. Richard Robb at the Biomedical Imaging Resource, Mayo Foundation (Rochester, MN). "This interaction with three-dimensional (3-D) imagery enhances the scientific and medical value of the data produced by medical imaging systems," he adds.
Such VR systems allow images from CT, MRI, and PET systems to be visualized as 3-D models. As a clinical aid to the preoperative process, physicians can interact with these models before attempting a surgical procedure. Sophisticated VR systems augment 3-D data with head-up displays, magnetic position trackers, and data gloves for more realistic virtual operations. At the leading edge, such VR systems are being integrated with robotic systems for surgical education, catheter placement, skull drilling, and bone repositioning.
Currently, many image-guided-surgery procedures are limited to cranial surgery. In this approach, the patient`s head is precisely positioned and secured by a metallic frame. After a CT scan is performed, points of reference on the head frame enable the physician to calculate a trajectory to any intracranial target visible on the scan. This procedure does not require direct vision of the target. However, the surgeon cannot check the exact position of the tip of an inserted surgical probe because brain tissue shift can occur.
"It does not take long for neurosurgeons to imagine the advantages that robots offer in performing such operations," says Marc Epitaux, an engineer with the DMT-ISR Group for surgical robotics and instrumentation, EPFL (Lausanne, Switzerland). Robots are more precise and can work faster and in reduced working environments, claims Epitaux. By functioning inside a CT scanner, robots such as EPFL`s Minerva can be incrementally and precisely moved during brain surgery. This capability allows the CT scanner to obtain new images at any stage, check the position of the instrument inside the brain, detect and observe cysts or hemorrhages in real time, and control a tumor biopsy.
By restricting the physician`s tasks to selections of the entry point into the skull and the target point, the Minerva robot calculates the transformation of the corresponding coordinates, thereby minimizing human error. Under the physician`s command, the robot cuts the skin, perforates the skull, penetrates the protective tissue layers covering the brain, and sends a probe toward the determined target point in the brain to perform a tissue biopsy or other procedure.
Still in the research-and-development stage, few such vision-guided surgery robots are now in use. However, several experimental procedures have been performed on animal cadavers and live humans. Unfortunately, despite the continuous advances in the development of robotic-guided machine-vision systems, the development of robot surgeons is likely to be limited by ethical and moral considerations.