Handheld micromanipulator for robot-assisted stapes footplate surgery

Abstract

Stapes footplate surgery is complex and delicate. This surgery is carried out in the middle ear to improve hearing. High accuracy is required to avoid critical tissues and structures near the surgical worksite. By suppressing the surgeon's tremor during the operation, accuracy can be improved. In this paper, a fully handheld active micromanipulator known as Micron is evaluated for its feasibility for this delicate operation. An ergonomic handle, a custom tip, and a brace attachment were designed for stapes footplate surgery and tested in a fenestration task through a fixed speculum. Accuracy was measured during simulated surgery in two different scenarios: Micron off (unaided) and Micron on (aided), both with image guidance. Preliminary results show that Micron significantly reduces the mean position error and the mean duration of time spent in specified dangerous zones.

Fig. 1

(a) Micron device outfitted with end-effector similar to (b) micropicks currently used in stapedectomy surgery. (c) Brace attachment and non-perpendicular ergonomic handle developed for otologic Micron.

Fig 2

The selected procedure for the experiment tries to mimic the motions of the stapes fenestration procedure. The experiment steps are : (a) make one first fenestration in the first hole, (b) move from tip of Micron from the first hole to the second one, following the line defined by the center of the first and the second hole and finally (c) make a second fenestration in the second hole.

Fig. 3

Work environment: (a) Micron, (b) microscope, (c) stereo cameras, (d) ASAP optical trackers, and (e) augmented reality projector (f) HD monitor.

Fig. 4

Visual cues injected into the surgical microscope. (a) A green circle marks goal position and depth. (b) A blue circle shows the null position of the micromanipulator. (c) Yellow lines delimit virtual fixture work space. They change depending on tip position: (1) Aiming at the first hole, (2) moving between holes with (d) a green line to show the desired path, and (3) aiming the second hole. If the tool is too high, the blue circle grows larger than the green circle by an amount proportional to the Z-depth error; thus visual cues show guidance in 3D.

Fig 5

2D Positioning error, aided (Micron on) vs. unaided (Micron off). An asterisk denotes statistically significant results (p < 0.05).

Fig. 6

Fenestration performance in a penetrations task: (left) Unaided traced path (with Micron off); (right) Aided traced path (with Micron on) with virtual fixtures.

Fig 7

Examples of intrusion into forbidden region during the procedure, as indicated by output near 1.5 V. (top) Unaided. (bottom) Aided.

Fig. 8

Forbidden areas intrusion time, aided vs. unaided, both cases with image guidance. An asterisk denotes statistically significant results (p < 0.05).