Sandie CHAMBERS^1,2, Jane Ladlow²

¹Glasgow University, Glasgow, United Kingdom. ²Cambridge University, Cambridge, United Kingdom

Objectives

This study establishes a reproducible CT‑based workflow that combines volumetric analysis with intranasal tissue and lesion‑site mapping to enable objective BOAS phenotyping across brachycephalic breeds and functional grades (0–3). A standardised three‑compartment segmentation model was applied to quantify the volume of the turbinate–mucosa complex and to describe rostral, mid‑cavity, and caudal tissue configurations, examining how these vary between and within breeds across the BOAS grading spectrum.

Methods

The study population included 8 brachycephalic dog breeds presented for upper airway assessment with complete CT imaging and documented BOAS functional grades (0–3). Retrospective CT scans were anonymised and classified by breed, sex, and BOAS grade. A modular segmentation workflow separated the turbinate–mucosa complex, nasal airspace, and other intranasal soft tissues using strict anatomical rules. Rostral, mid‑cavity, and caudal slices were used for quality assurance, and 3D reconstructions were generated. Volumetric measurements and mapped intranasal tissue patterns were documented across breeds and correlated with BOAS functional grades.

Results

Volumetric increases in the turbinate mucosa complex and reductions in the nasal airspace showed a clear progression between functional grades 0 to 3. Grade 3 dogs demonstrated marked mid‑cavity tissue crowding and frequent rostral aberrant turbinates. Bulldogs showed extensive ventral conchal enlargement, while pugs displayed pronounced ethmoturbinate involvement. The workflow was reproducible and enabled consistent separation of turbinate tissue from adjacent structures.

Statement: Impact/ Clinical Significance

This study introduces the first volumetric and tissue‑mapping method for assessing intranasal obstruction in BOAS, demonstrating clear correlations between intranasal tissue volume, tissue configuration, and BOAS grade. These findings support an imaging‑based framework for breed‑specific risk assessment, lesion site identification, refinement of BOAS grading, and improved surgical planning.