Filing activity in this dataset is concentrated among a group of orthopaedic and surgical-robotics companies, including Smith & Nephew entities, Stryker, Medtronic Navigation, Globus Medical and Intuitive Surgical Operations, alongside academic and research filers such as Mayo Foundation. Several of the most active historical filers show zero filings in the latest recorded year, which may reflect either a genuine pause or the roughly 18-month lag between filing and publication. No single assignee dominates the full 400-family set outright; the pattern is concentration among a handful of established names with a long tail of smaller filers behind them.

Based on this dataset, filing volume peaked in 2020 at 66 families and has declined since, with the 2022 midpoint already down to 39. That points to a category that has passed its peak filing rate rather than one still accelerating, though the most recent one to two years are understated because of publication lag and should not be read as a hard drop to near zero. The more reliable read is that core registration and tracking claims are increasingly staked out, pushing new filers toward narrower clinical applications.

The core classifications are A61B34/20, A61B90/00 and A61B34/10, covering surgical navigation systems, instrument tracking and computer-assisted surgery. Beyond that core, G06T (image data processing and generation) appears in 109 of 400 families, making it by far the largest adjacent classification — well ahead of A61F implants at 19 or A61M fluid devices at 16. Anyone drafting claims in registration, AR overlay or 3D planning should check G06T-classified prior art specifically, not just the A61B core.

The classification data shows comparatively thin coverage in areas like magnetic instrument tracking, sterilisable navigation optics, and AR-overlay claims tied to fluid-delivery or spinal-cord-therapy applications, relative to the dense core registration and tracking claims. Geographic white space also exists: China and Canada each show only 15 filings against 190 in the US, so claims drafted with those jurisdictions specifically in mind face a thinner prior-art wall. Neither observation guarantees patentability, but both indicate where a first claim is less likely to collide with existing filings.

The most-cited records in this dataset, including US20160191887A1 and US20180240237A1, are also among the older filings, which is typical: citation counts inside any searched corpus favour records that have had more years to accumulate references. A high citation count is a reasonable signal of historical influence on how the field's claims were framed, but it is not a reliable signal of which technology is commercially dominant today. Recent filings with few citations so far may still represent the current state of the art; they simply have not had time to be cited.