AI‐Enabled Mucus Segmentation in Nasal Endoscopy with State Space and Attention‐Based Modeling

Dipesh Gyawali, Jonathan Bidwell, Elena Karras, Thomas J Mundy, Kayla E Baker, Elena D Bartolone, Dhara Patel, Akio Fujiwara, Edward D McCoul · 2026

Read AI‐Enabled Mucus Segmentation in Nasal Endoscopy with State Space and Attention‐Based Modeling by Dipesh Gyawali, Jonathan Bidwell, Elena Karras, Thomas J Mundy, Kayla E Baker, Elena D Bartolone, Dhara Patel, Akio Fujiwara, Edward D McCoul (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13261536.txt.

Description

Dipesh Gyawali and colleagues present a 2026 study of an AI model that segments mucus in nasal endoscopy images. Outlining mucus helps diagnose sinusitis, but thin layers, shiny secretions, glare, and overexposure defeat existing models. Their model, SUM-MucusNet, was built to handle these challenges for clinicians.

The team trained the model on 3,486 annotated images from 454 patients and compared it with expert observations and existing models. It reached a Dice coefficient of 0.71, a modest but statistically supported improvement over the next-best model (p = 0.027). In poor-quality images it correctly segmented 89.3% of mucus regions. Boundary detection improved for irregular mucus patterns, and false positives from glare fell. The system ran at 17.3 frames per second, fast enough for real-time use. The authors aim at standardized, quantitative mucus assessment over time.

Catalog record

CREATORDipesh Gyawali, Jonathan Bidwell, Elena Karras, Thomas J Mundy, Kayla E Baker, Elena D Bartolone, Dhara Patel, Akio Fujiwara, Edward D McCoul
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE27326
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13261536

Browse the full library · About the Tower of Babel