The ranking covers 100 assignees across 664 records, and it is heavily front-loaded: the leading assignee alone holds 222 records, and the top 5 combined account for 391 records, or 58.9% of everything in scope. The leaders are the CT-system manufacturers you would expect — Philips, GE, Canon, Siemens Healthineers and their group entities — plus Mayo Foundation as the one academic-medical presence in the top tier. Below the top 10, which together hold 72.4% of records, filing activity fragments quickly into a long tail of few- or single-filing entrants.

Filing volume peaked at 75 records in 2017 and has since eased; the clearest complete-year comparison shows 2021's 49 records falling to 29 by 2024, a 41% decline. That is a real trend for years that are fully counted, but publication typically lags filing by around 18 months, so 2025 and 2026 figures in any such dataset are still incomplete and should not yet be read as confirmation of further decline. A fair read is that the field has already been through a filing wave and cooled from its peak, not that it has stopped.

Material decomposition is the process of using spectral (multi-energy) CT data to separate an image into component materials — commonly iodine contrast versus soft tissue or bone — rather than producing a single density-based grayscale image. Patent claims in this space cover the acquisition method (how dual- or multi-energy data is captured), the decomposition algorithm itself, and calibration routines that keep iodine quantification accurate and control noise amplification. In this dataset, 62.3% of records carry an A61B (diagnosis and surgery) classification and 38.6% carry G06T (image processing), showing that most claims sit at the clinical-application and algorithmic layers.

Relative to the field's dominant A61B and G06T claim volume, classifications tied to AI-model-based calibration (G06N, 5.4% of records), data recognition and presentation (G06K, 5.3%), healthcare informatics integration (G16H, 4.2%) and X-ray source technique (H05G, 3.2%) are comparatively thin. That does not mean these areas are unclaimed, but claim density there is much lower than in core decomposition and image-processing claims, which is where a narrower, well-drafted first claim has more room to stand on its own prior art rather than colliding with incumbent families.

The most-cited records in this dataset date back to the mid-2000s and 2010s, including filings on identification and imaging of specific materials and image-based material decomposition methods, some cited over 90 times. High citation counts reflect the fact that older documents inside a searched corpus have simply had more time to accumulate references, so they signal historical influence on how the field's claims were framed rather than current technical importance. A newer, less-cited filing can still be more relevant to a present-day freedom-to-operate check.