Methods for Madigan Endoscopic Sinus Surgery Simulator v.2.0
 
Tasks Performed (Methods)

Model 1: The novice/abstract model consisted of only the skin of the face and the entrance to the nasal cavity. A 3D grid pattern replaced the sinus anatomy to provide depth of field during the three subtasks: Navigation, Injection and Dissection. Training aids were used to guide the subjects through the subtask. Navigation training aids consisted of virtual hoops and Injection training aids consisted of virtual targets in space.

During Navigation the subjects maneuvered the instrumented endoscope through four sets of virtual hoops. The paths of the hoops represented three passes (sets two and three combined are one pass) commonly taken before the surgery begins to gain familiarity with the patient's anatomy and to allow cleaning of the areas of interest.

Injection consisted of coordinating both the endoscope and instrumented forceps within the environment to inject five targets oriented obliquely in space. The instrumented forceps controlled a virtual needle for this subtask. The placement of the targets in space reflected the common areas of injection of a vasoconstrictor during a Maxillary Antrostomy.

During Dissection the subjects were also required to use both the endoscope and the instrumented forceps. The subtask consisted of a tool-specific obstacle and a dissectable hazard surrounding the obstacle. The subjects were to dissect these obstacles with pre-selected virtual tools and avoid the hazards. The instrumented forceps represented each of the tools most commonly used in the procedure: Left side-bitter, Up-bitter, Microdebrider, Sickle Knife, Router, Left Antrum Punch, Round Burr, Straight-bitter, Left Scissors, Bent Router, and Circular Antrum Punch.

Navigation through the four sets of hoops, injection of the five targets, and dissection of each of the obstacles was required for a complete score. If the hazards surrounding the Dissection obstacles were hit, their score was decreased by a percentage related to the amount of the hazard dissected. Digitized audio cues were given for each hoop negotiated in Navigation, for each target Injected, for percentages completed of each sphere during Dissection, for interaction with hazards, and for completion of each subtask.

Model 2: The intermediate model was composed of the Navigation hoops and Injection targets from Model 1 overlaid within a virtual anatomical model of the paranasal sinus. Injection and dissection followed the protocol for a Total Ethmoidectomy. Text labels were overlaid on all anatomical structures with which the subject interacted.

Navigation through the four sets of hoops, injection of the five targets, medialization of the Middle Turbinate, dissection of the five anatomical structures, removal of two bone fragments placed in the Uncinate process and removal of three bone fragments placed in the Bulla ethmoidalis were required for a complete score. The hazards in this model were: Anterior Ethmoid Artery, Carotid Artery, Lamina Papyracea, Opthalmic Artery, Optic Nerve, Periorbital and Periorbital Fat. Audio cues were given for each hoop negotiated in Navigation, for each target hit in Injection, for each bone fragment removed, for interaction with hazards, for percentages completed for each anatomical structure in Dissection, and for completion of each subtask.

Model 3: The advanced model was composed of an anatomical model only. Subjects were expected to perform the three subtasks without the training aids and to follow the protocol for a Maxillary Antrostomy, Anterior Ethmoidectomy and removal of the Agar Nasi Cell. Three polyps were added superior/anterior to the Bulla ethmoidalis.

During Navigation the subject was required to perform the three passes in the same order as in Models 1 and 2: inferior pass along the floor of the nose to the Nasopharynx, followed by an intermediate pass medial to the Middle Turbinate towards the upper aspect of the Nasopharynx and Sphenoid Ostium, then rolling under the Middle Turbinate to inspect the Ostial Meatal Complex, and finally the superior pass medial to the root of the Middle Turbinate towards the Sphenoethmoidal Recess.

During Injection, the subject was cued only by the amount of blanching (whitening) of the virtual tissue as to whether more vasoconstrictor was needed.

Dissection followed the same procedure as Model 2. Navigation through the three passes, injection of the areas of interest, medialization of the Middle Turbinate, dissection of the five anatomical structures, dissection of the three polyps, removal of two bone fragments placed in the Uncinate process and removal of three bone fragments placed in the Bulla ethmoidalis were required for a complete score. Hazards were the same as Model 2. Audio cues were given only at the end of each of the Navigation passes, for removal of bone fragments and for percentages completed in Injection and Dissection.

last updated: 10/12/98