Claudia Scherl1, David Männle2, Nicole Rotter2, Jürgen Hesser3, Jan Stallkamp3, Tobias Balkenhol2, Lena Huber2, Benedikt Kramer2, Anne Lammert2, Annette Affolter2. 1. Department of Otorhinolaryngology, Head and Neck Surgery, University Medical Center Mannheim, Medical Faculty Mannheim, Heidelberg University, Theodor-Kutzer-Ufer 1-3, 68167, Mannheim, Germany. claudia.scherl@umm.de. 2. Department of Otorhinolaryngology, Head and Neck Surgery, University Medical Center Mannheim, Medical Faculty Mannheim, Heidelberg University, Theodor-Kutzer-Ufer 1-3, 68167, Mannheim, Germany. 3. Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University, Theodor Kutzer-Ufer 1-3, 68167, Mannheim, Germany.
Abstract
PURPOSE: Augmented Reality can improve surgical planning and performance in parotid surgery. For easier application we implemented a voice control manual for our augmented reality system. The aim of the study was to evaluate the feasibility of the voice control in real-life situations. METHODS: We used the HoloLens 1® (Microsoft Corporation) with a special speech recognition software for parotid surgery. The evaluation took place in a audiometry cubicle and during real surgical procedures. Voice commands were used to display various 3D structures of the patient with the HoloLens 1®. Commands had different variations (male/female, 65 dB SPL)/louder, various structures). RESULTS: In silence, 100% of commands were recognized. If the volume of the operation room (OR) background noise exceeds 42 dB, the recognition rate decreases significantly, and it drops below 40% at > 60 dB SPL. With constant speech volume at 65 dB SPL male speakers had a significant better recognition rate than female speakers (p = 0.046). Higher speech volumes can compensate this effect. The recognition rate depends on the type of background noise. Mixed OR noise (52 dB(A)) reduced the detection rate significantly compared to single suction noise at 52 dB(A) (p ≤ 0.00001). The recognition rate was significantly better in the OR than in the audio cubicle (p = 0.00013 both genders, 0.0086 female, and 0.0036 male). CONCLUSIONS: The recognition rate of voice commands can be enhanced by increasing the speech volume and by singularizing ambient noises. The detection rate depends on the loudness of the OR noise. Male voices are understood significantly better than female voices.
PURPOSE: Augmented Reality can improve surgical planning and performance in parotid surgery. For easier application we implemented a voice control manual for our augmented reality system. The aim of the study was to evaluate the feasibility of the voice control in real-life situations. METHODS: We used the HoloLens 1® (Microsoft Corporation) with a special speech recognition software for parotid surgery. The evaluation took place in a audiometry cubicle and during real surgical procedures. Voice commands were used to display various 3D structures of the patient with the HoloLens 1®. Commands had different variations (male/female, 65 dB SPL)/louder, various structures). RESULTS: In silence, 100% of commands were recognized. If the volume of the operation room (OR) background noise exceeds 42 dB, the recognition rate decreases significantly, and it drops below 40% at > 60 dB SPL. With constant speech volume at 65 dB SPL male speakers had a significant better recognition rate than female speakers (p = 0.046). Higher speech volumes can compensate this effect. The recognition rate depends on the type of background noise. Mixed OR noise (52 dB(A)) reduced the detection rate significantly compared to single suction noise at 52 dB(A) (p ≤ 0.00001). The recognition rate was significantly better in the OR than in the audio cubicle (p = 0.00013 both genders, 0.0086 female, and 0.0036 male). CONCLUSIONS: The recognition rate of voice commands can be enhanced by increasing the speech volume and by singularizing ambient noises. The detection rate depends on the loudness of the OR noise. Male voices are understood significantly better than female voices.
Authors: Claudia Scherl; Johanna Stratemeier; Nicole Rotter; Jürgen Hesser; Stefan O Schönberg; Jérôme J Servais; David Männle; Anne Lammert Journal: ORL J Otorhinolaryngol Relat Spec Date: 2021-03-30 Impact factor: 1.538
Authors: Claudia Scherl; Johanna Stratemeier; Celine Karle; Nicole Rotter; Jürgen Hesser; Lena Huber; Andre Dias; Oliver Hoffmann; Philipp Riffel; Stefan O Schoenberg; Angela Schell; Anne Lammert; Annette Affolter; David Männle Journal: Eur Arch Otorhinolaryngol Date: 2020-09-10 Impact factor: 2.503
Authors: Cristina M Morales Mojica; Jose D Velazco-Garcia; Eleftherios P Pappas; Theodosios A Birbilis; Aaron Becker; Ernst L Leiss; Andrew Webb; Ioannis Seimenis; Nikolaos V Tsekos Journal: J Digit Imaging Date: 2021-05-23 Impact factor: 4.903
Authors: Jun Liu; Subhi J Al'Aref; Gurpreet Singh; Alexandre Caprio; Amir Ali Amiri Moghadam; Sun-Joo Jang; S Chiu Wong; James K Min; Simon Dunham; Bobak Mosadegh Journal: PLoS One Date: 2019-07-01 Impact factor: 3.240
Authors: Fabio Cofano; Giuseppe Di Perna; Marco Bozzaro; Alessandro Longo; Nicola Marengo; Francesco Zenga; Nicola Zullo; Matteo Cavalieri; Luca Damiani; Daniya J Boges; Marco Agus; Diego Garbossa; Corrado Calì Journal: Front Surg Date: 2021-03-30