Comparison of filter-based and software-based image enhancement systems for endoscopic swallowing diagnostics: a...

Area of Science:

• Medical Imaging
• Gastroenterology
• Otolaryngology

Background:

• Flexible endoscopic evaluation of swallowing (FEES) is a crucial diagnostic tool for swallowing disorders.
• Image enhancement technologies aim to improve visualization during FEES procedures.

Purpose of the Study:

• To compare the effectiveness of hardware filtering (Narrow Band Imaging, NBI) and software filtering (Professional Image Enhancement Technology, PIET) in enhancing food dye visibility during FEES.
• To assess the impact of these technologies on subjective image quality, color intensity, and contrast.

Main Methods:

• A head-to-head comparison of NBI and PIET using 190 videos of food dyes at various dilutions (1:10 to 1:100,000).
• Evaluation of maximum visible dilution by four raters across all colors.
• Analysis of 16 video snippets (eight colors, 1:10 dilution) by 14 blinded raters, including novices.

Main Results:

• Both NBI and PIET demonstrated similar hues for red, yellow, and their secondary colors.
• PIET was preferred for red and purple; NBI for yellow, green, and orange.
• PIET was overwhelmingly preferred for blue, black, and white. Raters achieved 100% agreement on maximum visible dilution, with both systems enhancing visibility tenfold.

Conclusions:

• Both NBI and PIET are effective in enhancing food dye visibility in FEES.
• Each technology offers distinct advantages, with the choice depending on user preference and system availability.
• Further research may explore optimal settings for specific clinical applications.