Introduction
The project is aimed at increasing the skill level of trainee surgeons who undertake colonoscopy operations. The normal procedure of performing colonoscopy involves physically screening patients using an optical endoscopic tool. The procedure takes about half a day and involves bowel flush preparation, sedation, and potential complications.
The patient's colon is inflated with air and the endoscope is navigated through the colon to inspect for (and remove) suspect polyps. Complications with the physical positioning and path of the patient's colon can prevent the surgeon from investigating the entire length of the colon. For these patients the typical follow up process involves performing Computed Tomography (CT) Colonography (virtual colonscopy). CT Colonography involves the patient taking a bowel preparation to flush the colon and in some acquisition protocols, ingestion of a contrast agent. The scanning procedure takes approximately 15 minutes, where CT scans of the patient are taken in two positions while the colon is inflated. The CT images of the colon are segmented and polyps identified.
Our custom designed haptic device allows interaction with the virtual colon using a real colonoscope.
Neither optical endoscopic colonoscopy nor CT Colonography are fully covered under the Australian Medicare rebate. Patients who are able to pay for CT Colonography may undertake this procedure instead of optical colonoscopy. However there is currently a 27% rate of polyp detection, so the cost of having to undergo two operations is a barrier to some patients who undertake this operation.
Aim
By improving the skill level of of optical colonoscopy the cost to the patient (and Australian health care system) will be reduced by not having to pay for CT Colonography. The aim of this project is to build a patient specific surgical training system to increase the success rate of optical colonoscopy operations. The surgical training system will allow the surgeon to navigate through a segmented image of the patient's colon, and provide haptic and visual feedback through a mechanical model of the colon. By building a database of difficult colonoscopy operations surgeons will be able to improve their skills for future operations.
Our real-time OpenGL rendering engine supports dynamic effects including red-out, lens flushing, mucous and surface pathologies.
