Peak Woo1, Peter Baxter2. 1. Department of Otolaryngology Head and Neck Surgery, Icahn School of Medicine, New York, New York. Electronic address: peakwoo@peakwoo.com. 2. Department of Otolaryngology Head and Neck Surgery, Icahn School of Medicine, New York, New York.
Abstract
OBJECTIVE: High-speed video (HSV) imaging of vocal fold vibration has been possible only through the rigid endoscope. This study reports that a fiberscope-based high-speed imaging system may allow HSV imaging of naturalistic voicing. MATERIALS AND METHODS: Twenty-two subjects were recorded using a commercially available black and white high-speed camera (Photron Motion Tools, 256 × 120 pixel, 2000 frames per second, 8 second acquisition time). The camera gain is set to +6 db. The camera is coupled to a standard fiber-optic laryngoscope (Olympus ENF P-4) with a 300-W Xenon light. Image acquisition was done by asking the subject to perform repeated phonation at modal phonation. Video images were processed using commercial video editing and video noise reduction software (After effects, Magix, and Neat Video 4.1). After video processing, the video images were analyzed using digital kymography (DKG). RESULTS: The HSV black and white video acquired by the camera is gray and lacks contrast. By adjustment of image contrast, brightness, and gamma and using noise reduction software, the flexible laryngoscopy image can be converted to video image files suitable for DKG and waveform analysis. The increased noise still makes edge tracking for objective analysis difficult, but subjective analysis of DKG plot is possible. CONCLUSIONS: This is the first report of HSV acquisition in an unsedated patient using a fiberscope. Image enhancement and noise reduction can enhance the HSV to allow extraction of the digital kymogram. Further image enhancement may allow for objective analysis of the vibratory waveform.
OBJECTIVE: High-speed video (HSV) imaging of vocal fold vibration has been possible only through the rigid endoscope. This study reports that a fiberscope-based high-speed imaging system may allow HSV imaging of naturalistic voicing. MATERIALS AND METHODS: Twenty-two subjects were recorded using a commercially available black and white high-speed camera (Photron Motion Tools, 256 × 120 pixel, 2000 frames per second, 8 second acquisition time). The camera gain is set to +6 db. The camera is coupled to a standard fiber-optic laryngoscope (Olympus ENF P-4) with a 300-W Xenon light. Image acquisition was done by asking the subject to perform repeated phonation at modal phonation. Video images were processed using commercial video editing and video noise reduction software (After effects, Magix, and Neat Video 4.1). After video processing, the video images were analyzed using digital kymography (DKG). RESULTS: The HSV black and white video acquired by the camera is gray and lacks contrast. By adjustment of image contrast, brightness, and gamma and using noise reduction software, the flexible laryngoscopy image can be converted to video image files suitable for DKG and waveform analysis. The increased noise still makes edge tracking for objective analysis difficult, but subjective analysis of DKG plot is possible. CONCLUSIONS: This is the first report of HSV acquisition in an unsedated patient using a fiberscope. Image enhancement and noise reduction can enhance the HSV to allow extraction of the digital kymogram. Further image enhancement may allow for objective analysis of the vibratory waveform.
Authors: Patrick Schlegel; Melda Kunduk; Michael Stingl; Marion Semmler; Michael Döllinger; Christopher Bohr; Anne Schützenberger Journal: PLoS One Date: 2019-04-22 Impact factor: 3.240