Many up-and-coming therapeutic protocols in ophthalmology are technically difficult, near/beyond the limits of human ability, and are being attempted by only a few surgeons. For example, subretinal injection of stem cells or gene therapies requires placement of a fine cannula in the subretinal space, holding that position steady for ∼90 s to inject a bleb of fluid. Surgeon hand tremor places a limit on achievable precision. A wide variety of robot-assisted surgical systems have been proposed to improve the precision of eye surgery. However, there has not been much consideration of patient head motion, which is common among patients undergoing eye surgery under monitored anesthesia, a.k.a. conscious sedation, which makes a patient calm and somewhat sleepy during a surgery, but the patient may still be awake. Head motion in this state is due to breathing, talking, snoring, and other (in)voluntary motions of the patient. Movement must be compensated by the surgeon, to the best of their ability, to avoid complications. Any clinical robotic system must deal with patient motion. In this paper, we introduce a noninvasive head-mounting concept, based on a modified radiotherapy immobilization system, which enables one or more robots to be mounted semi-rigidly to a patient’s head to passively compensate for patient motion, and we perform an experiment that highlights the potential benefits of this form of head mounting.