Infrastructure for Image-Guided Surgery at UIHC

John Haller, PhD, Timothy Ryken, MD and Michael Vannier, MD

        Image guided surgery at UIHC consists of an integrated environment for seamless acquisition, visualization, processing and storage of images and associated information. The infrastructure to support image guided surgery integrates image acquisition, networks, presurgical planning, surgical navigation, and data archiving. Following is a pictoral description of the principal components of an integrated image guided surgery environment and its implementation at the University of Iowa Hospitals and Clinics.

Image guided neurosurgery  utilizes a computer workstation and optic guidance which localizes the instruments and patient’s anatomy in the 3D space of the operating room.  Virtual representations of the surgical instruments are projected onto the patient’s MRI or CT images displayed on a computer monitor.

        The computer monitor of a surgical navigation workstation displays the patient’s images that are used to guide the surgeon during surgery.  Magnetic resonance images and 3D reconstruction used in image-guided surgery show the brain tumor and zones for 1 mm (green sphere/circle) and 2 mm (yellow circle) accuracy.  Blue circles surround external skin markers used to register the virtual image space with the actual head of the patient,  so that the images can be used to guide the surgeon through the anatomy of the brain.

Surgical Decision Support:

Critical Information at the point of care

          Critical information can be added to the images used for surgical guidance in the operating room.  For example, MRI images showing brain activity can be used to depict the region of the brain responsible for hand movement.  A tumor (white, enhanced object) near the motor area of the brain is shown along with the trajectory of a surgical instrument (green line) used during surgery.  In addition, ultrasound images (lower right) acquired during surgery can be co-registered with MRI (lower left) and used to update the position of the tumor and surrounding anatomy.  These kinds of images will soon be available to surgeons at UIHC in a heads-up display (HUD).

               Images are transferred  from CT or MRI scanners (upper right) through the imaging network (2-5 minutes) to a workstation where surgical planning and 3D modeling is done (bottom right). After the patient is positioned for treatment in the operating room, calibration is done to place all of the pre-op images and surgical instruments into a common reference frame (bottom left). As the surgery proceeds, a navigation workstation is used to guide the procedure pre-operative images (upper left). The surgical navigation can also be augmented with real-time images from ultrasound or fluoroscopy, if needed.

               The network for image-guided surgery at UIHC includes connections from CT and MRI scanners to radiology image review workstations and a surgery-dedicated server.  Presurgical planning can be done on remote workstations using secure web-based protocols on a personal computer.    Finally, post-processed images can be transferred from the server directly to the O.R. surgical navigation workstations.

            At UIHC a storage area network (SAN) and Radiology Department PACS archive (27 terabytes) provide complementary image storage capabilities that meet the needs of routine clinical operations, research, and image guided surgery.  The SAN provides high speed connections to a RAID (Redundant Array of Independent Disks) mass storage device. Multiple workstations have simultaneous access through fiber optic connections to a fiber switch.  The origin of the data is transparent to the user who has instant access to the RAID, similar to a true multi-user system using internal disks.  This system can be greatly expanded with more RAIDs or other disks. Only the fiber switch limits the number of PCs or Unix computers that can be connected to the RAID.

                 After the images are transferred to the operating room the the patient is prepared for for image guided surgery which includes attaching a reference arc with light emitting diodes (LEDs).  The black arc pictured here is attached by a three-point clamp (Mayfield clamp) to the patient’s head.  Fiducial markers glued to the skin are localized by the surgical probe (upper left), which also has LEDs that are detected by an infrared sensor connected to the surgical navigation workstation.  The black line drawn on the patients head indicates the extent of the tumor boundaries and surgical incision planned using CT image guidance.

         3D images can be used during surgical navigation to visualize the brain’s cortical surface and veins (top row), as well as the location of the tumor in slices (middle left) or cut-away views (bottom row).

To find out more, click here to download a full pdf file of this article