Augmented reality in jaw reconstruction: Northwick Park’s HoloLens experience

A reported use in reconstruction

According to London North West University Healthcare NHS Trust (LNWH), a team at Northwick Park Hospital used HoloLens 2 augmented-reality glasses during reconstructive surgery for a patient with cancer. LNWH reported that the operation lasted 12 hours and involved removing diseased jawbone, then reconstructing the jaw using bone and attached blood vessels harvested from the lower leg.

For surgeons interested in technology and innovation, the reported case raises a focused question: how might a virtual anatomical model support visualisation during a complex reconstructive procedure? The distinction between describing a technology’s use and showing its effect on care is important when considering this account.

Bringing scans into view

LNWH reported that the team used a virtual model derived from preoperative scans and aligned it with the patient’s leg anatomy. According to the trust’s account, hand gestures and voice commands were used to position the model. The reported application therefore concerned visualising the lower-leg anatomy involved in obtaining tissue for the jaw reconstruction.

The relationship between that virtual model and the patient’s anatomy is a central point of interest. From the details available in LNWH’s report, it is not yet clear how alignment accuracy was assessed, whether the overlay required adjustment during the procedure, or how any mismatch was identified. Those questions are separate from the reported ability to manipulate and align a model.

Understanding the workflow

LNWH’s description places the glasses within an operation that combined removal of diseased jawbone with reconstruction using tissue from the lower leg. Although the trust reported a total operating time of 12 hours, that figure alone does not show whether augmented reality shortened or extended any part of the procedure. It is not yet clear how much of the operation involved use of the glasses.

Similarly, LNWH’s account of gesture and voice control describes an interaction method, rather than a measured effect on the surgical workflow. Details of preparation time, model creation, staff training and the division of tasks within the team are not yet clear. These remain areas for further reporting rather than grounds for assuming either an efficiency gain or an additional burden.

Interpreting the experience

The experience reported by LNWH may interest surgeons exploring how scan-derived models can be viewed alongside operative anatomy. Its relevance lies in the specific application described: aligning a virtual model with the leg during a procedure involving transfer of bone and attached blood vessels to reconstruct the jaw.

However, the reported use should not be read as evidence of improved precision, fewer complications or better recovery. From the information available in LNWH’s account, comparative results and longer-term patient outcomes are not yet clear. Nor is it clear how readily the same approach could be used across different patients, teams or reconstructive procedures.

Questions for future reporting

LNWH’s reported experience opens several avenues for discussion: how alignment is checked, how users interact with the display during different stages of surgery, and how any effects on workflow or outcomes might be measured. More detail on these points would help distinguish the practical experience of using an overlay from evidence about its wider value.

This article is an overview for information only; readers should consult the original sources and appropriate professional guidance.