Laryngeal Videoendoscopy Research Articles

Towards image-based laryngeal videostroboscopy using deep learning-enabled compressed sensing

Laryngeal videostroboscopy is an audio-mediated imaging technique allowing the visualization of vocal fold oscillation behavior: the audio signal is used to determine the fundamental frequency F0, which represents the vocal fold oscillation frequency. Knowing F0 allows to trigger the strobe illumination unit to provide a still image or slow-motion view of the vocal fold oscillation. However, this procedure involves several hardware components, noisy audio signals, and a chain of complex, error-prone algorithms that have to be orchestrated. We hypothesize that endoscopic images suffice to determine F0 with a view towards providing an alternative, image-based approach for estimating F0 during laryngeal videoendoscopy. In this study, we show that we are able to predict the relative glottal opening state to create sample points on the glottal area waveform, an endoscopic image-derived signal capable of deriving F0. As imaging frame rates from ordinary endoscopic cameras do not fulfill the Shannon–Nyquist criterion, we solve this problem with compressed sensing. We developed and evaluated the proposed approach using high-speed videoendoscopy (HSV) to simulate different, realistic low frame rates that are similar to those used in videostroboscopy. We show that we are able to predict F0 with over 95% accuracy using at most 75 sample points of a 600 ms long footage. Using endoscopic images and our algorithm only, we showcase that we can achieve a stroboscopic effect. This shows, that our proposed method in combination with the developed algorithm may be considered in the future to be integrated into clinical videostroboscopy.

Effect of a combination of butorphanol and detomidine on endoscopic assessment of laryngeal function in Thoroughbred yearlings.

To the authors' knowledge, the effect of a combination of butorphanol tartrate and detomidine hydrochloride on the laryngeal function and symmetry of the arytenoid cartilages in young Thoroughbred horses has not been examined. Therefore, this study aimed to examine the effects of administration of butorphanol and detomidine on left-to-right rima glottidis ratio (L:R RGR), cross-sectional area of the rima glottidis (CSARG) and grade of laryngeal function of Thoroughbred yearlings examined endoscopically at rest. Forty-six Thoroughbred yearlings underwent laryngeal video endoscopy, at rest, before and after sedation with butorphanol and detomidine. Three clinicians assigned grades of laryngeal function after reviewing the video recordings. The L:R RGR and CSARG were measured at the point of arytenoid cartilages' maximal abduction. Student's t-test was used to compare the L:R RGR and CSARG. Cohen's kappa (κ) was used to compare the intra- and interobserver agreements. Spearman's rank order was used to assess agreement between CSARG and grade of laryngeal function. Sedation had no effect on L:R RGR and CSARG. Mean intraobserver agreement of grade of laryngeal function of unsedated yearlings was 93.1%, and that of sedated yearlings was 92.9%. Mean interobserver agreement of grade of laryngeal function of unsedated yearlings was 92.8%, and that of sedated yearlings was 92.7%. Correlation between CSARG and laryngeal function grade was significant for both groups (p<0.001). Sedating Thoroughbred yearlings with detomidine and butorphanol, before endoscopically examining the larynx, did not significantly affect the grade of laryngeal function.

Clinical Images: Severe proteinase 3 antineutrophil cytoplasmic antibody glomerulonephritis temporally associated with Sinovac Biotech's inactivated SARS-CoV-2 vaccine.

Mass vaccination campaigns to prevent coronavirus disease 2019 (COVID-19) are occurring worldwide; not surprisingly, reports of rare postvaccination immune-mediated adverse events, as seen with the influenza vaccination (1), also started to surface, including cases of antineutrophil cytoplasmic antibody (ANCA)–associated vasculitis (AAV) after receipt of the Pfizer-BioNTech and Moderna COVID-19 vaccines (2, 3). Here, we report a case from Brazil of new-onset AAV temporally associated with inactivated COVID-19 vaccination. The patient, a previously asymptomatic 78-year-old White woman with a medical history of odontogenic maxillary sinusitis resolved after sinus surgery (in 2011), presented with asthenia, low-grade fever, and a mild dry cough 2 weeks after receiving Sinovac Biotech’s COVID-19 vaccine (CoronaVac) on March 12, 2021 (second shot on April 5, 2021). Over the weeks, in addition to persistent fever, the patient experienced worsening of fatigue, weight loss, night sweats, and renal dysfunction. Of note, various antibiotics were empirically used until she underwent a kidney biopsy on June 22. At that time, she had leukocytosis, anemia, thrombocytosis, elevated inflammatory biomarkers and serum creatinine (peak at 7.3 vs 0.8 mg/dl on May 1, 2021), cytoplasmic ANCA 1:20 by indirect immunofluorescence (reference: negative), an anti-proteinase-3 level of 19 U/ml by fluorescent-enzyme immunoassay (positive >3 U/ml), a normal complement level, polyclonal hypergammaglobulinemia, and active urinary sediment. The diagnostic workup was negative for COVID-19 (reverse transcription–polymerase chain reaction), antinuclear antibodies (ANA), rheumatoid factor, cryoglobulins, anti-myeloperoxidase, anti-phospholipids, anti-Ro, anti-La, anti-Smith (Sm), anti–double-stranded DNA, anti-ribonucleoprotein (RNP), anti–glomerular basement membrane, HIV, and hepatitis B and C. Computed tomography (head, chest, paranasal sinuses, and abdomen), echocardiography, and nasal and laryngeal videoendoscopy findings were either normal or nonspecific. Urine and blood culture results were negative. The result of a kidney biopsy (14 glomeruli yielded) was compatible with necrotizing (9 of 14 glomeruli) and crescentic (cellular crescents in 6 of 14 glomeruli) glomerulonephritis. The renal biopsy specimen shows a cellular crescent and break of the glomerular basement membrane (periodic acid–Schiff stain, original magnification ×40) (A), break in Bowman’s capsule (Jones stain, original magnification ×40) (B), an acute tubular injury (hematoxylin–eosin stain, original magnification ×40) (C), and fibrinoid necrosis in the glomerulus capillary loop (hematoxylin–eosin stain, original magnification ×40) (D). Renal immunostaining results were negative for C1q; C3; immunoglobulins A, M, and G; and light chains κ and λ. The patient received high-dose corticosteroids, pulse cyclophosphamide, and needed hemodialysis for 12 weeks. She is now controlled on low-dose corticosteroids and mycophenolate mofetil but with chronic renal impairment. To our knowledge, this is the first report of AAV possibly triggered by an inactivated COVID-19 vaccine. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Recurrent laryngeal nerve injury after thyroid surgery

Relevance: Thyroid surgery is dangerous and can lead to damage RLN. The frequency of nerve damage varies widely and this is due to the fact that not always immediately after surgery and subsequent examination of the larynx. This leads to late start of treatment. Purpose of this research is estimate the frequency of transient and permanent immobility of the vocal folds after thуroid surgery. Materials and methods: It was prospective cohort monocenters investigate. 1364 patients were examined. We developed “voice passport of the patient” to improve the diagnosis and treatment. It included data of examination and treatment. All patients in the preoperative and postoperative periods were examination: laryngoscopy, laryngeal videoendoscopy, voice spectral analysis and Voice Handicap Index (VHI-30 modified version) by Jacobcon. Results and discussion: Postoperative disorders of motion vocal folds were in 132 (9.7%) patients. Recovery of motion vocal folds were observed in 123 patients, which were 93.2% of the total number of cases of damage RLN (132) and 9.1% of the total number of surgical interventions (1364). Lack of recovery of motion vocal folds more than one year (laryngeal paralysis) was observed in 9 patients, which was 6,8% of the total number of cases of damage RLN (132) and 0.6% of the total number of surgical interventions (1364). Conclusions: Studying the frequency of paresis and paralysis of the larynx depending on the type of surgery and predicting the risks of PGN depending on the scope of the intervention before its implementation allows to reduce the number of postoperative complications and optimize management tactics in cases where nerve injury can’t be avoided.

BAGLS, a multihospital Benchmark for Automatic Glottis Segmentation

Laryngeal videoendoscopy is one of the main tools in clinical examinations for voice disorders and voice research. Using high-speed videoendoscopy, it is possible to fully capture the vocal fold oscillations, however, processing the recordings typically involves a time-consuming segmentation of the glottal area by trained experts. Even though automatic methods have been proposed and the task is particularly suited for deep learning methods, there are no public datasets and benchmarks available to compare methods and to allow training of generalizing deep learning models. In an international collaboration of researchers from seven institutions from the EU and USA, we have created BAGLS, a large, multihospital dataset of 59,250 high-speed videoendoscopy frames with individually annotated segmentation masks. The frames are based on 640 recordings of healthy and disordered subjects that were recorded with varying technical equipment by numerous clinicians. The BAGLS dataset will allow an objective comparison of glottis segmentation methods and will enable interested researchers to train their own models and compare their methods.

A bitless bridle does not limit or prevent dynamic laryngeal collapse.

Bits have often been incriminated as a cause of upper respiratory tract obstruction in horses; however, no scientific studies are available to confirm or refute these allegations. Clinical signs of dynamic laryngeal collapse associated with poll flexion (DLC) are induced when susceptible horses are ridden or driven into the bit. To determine whether use of Dr Cook's™ Bitless Bridle, instead of a conventional snaffle bit bridle, would reduce the severity of DLC in affected horses measured objectively using inspiratory tracheal pressures. Intervention study using each horse as its own control in a block randomised order. Nine Norwegian Swedish Coldblooded trotters previously diagnosed with DLC were exercised on two consecutive days using a standardised high-speed treadmill protocol with either a conventional bridle with a snaffle bit, or Dr Cook's™ Bitless Bridle. Head and neck position, rein tension, inspiratory tracheal pressure measurements, and laryngeal videoendoscopy recordings were obtained. A heart rate greater than 200bpm, and similar degrees of poll flexion/head height, had to be achieved in both bridles for the individual horse's data to be included for comparison. Seven horses' data met the inclusion criteria. The change in mean inspiratory tracheal pressure between free and flexion phases in the bitless bridle (-15.2±12.3cmH2 O) was significantly greater (P<.001) than in the snaffle bit bridle (-9.8±7.9cmH2 O). Mean inspiratory pressure during the free phase was significantly (P<.001) more negative with the snaffle bit bridle (-32.3±6.3cmH2 O), vs the bitless bridle (-28.5±6.9cmH2 O). Mean pressures in flexion phase, snaffle bridle (-42.1±10.8cmH2 O), vs bitless bridle (-43.7±15.6cmH2 O) where not significantly different between bridles (P=.2). Small sample size due to difficulty recruiting suitable clinical cases. This study could not provide any clear evidence that the effect of a snaffle bit in a horse's mouth influences the development or severity of DLC. Instead, head and neck angles induced by rein tension seem to be the key event in provoking DLC in susceptible horses.

High-technologic methods of laryngeal cancer diagnostics: a clinical case

Summary . The article observes the possibilities of high-technologic radiologic methods for diagnostics of laryngeal malignancy. Despite of large amount of publications, dedicated to the question of different radiologic imaging modalities application for laryngeal tumors the aspects of complex radiological examination and highly differentiated laryngeal tumors detection are not well described. Materials and methods . In our article there is the case of diagnostic process in 56-years old male patient complained on voice changes with the right vocal fold neoplasm detected in laryngeal videoendoscopy, which had ambiguous pattern in autofluorescence. The results of laryngeal videoendoscopy included differently colored zones from white ones, which corresponded to benign tissue, to blue and violet ones, which corresponded to malignancy. The patient was examined with contrast-enhanced computed tomography, contrast-enhanced magnetic resonance imaging with diffusion-weighted protocol and combined positron-emission and computed tomography with 18 F-fluorodeoxyglucose. Results. Computed tomography data showed the right vocal fold thickening without contrast agent uptake, on magnetic resonance images there was detected the irregular thickening of the same fold, also without contrast medium agent uptake. In positron-emission and computed tomography with 18 F-fluorodeoxyglucose a focal radiotracer hypermetabolism with highly increased standard uptake value was detected, which indicated malignancy. In addition distant metastases in thoracic lymph nodes were found. Pathomorphologic diagnosis of high-differentiated laryngeal cancer was established. Discussion. Highly differentiated laryngeal malignant tumors can imitate benign process on computed tomography and magnetic resonance imaging because of peculiar cell composition, but they are characterized with 18 F-fluorodeoxyglucose hyperfixation conditioned by glycolytical activity of malignant cells. Negative computed tomography and magnetic resonance imaging data in combination with positive positron-emission and computed tomography results may correspond to highly differentiated laryngeal tumor.

Demographic Profile and Risk Factors of Patients with Benign Vocal Fold Lesions Diagnosed through Laryngeal Videoendoscopy and Stroboscopy

Objective: To determine the prevalence of benign vocal cord lesions among Filipino patients in a tertiary institution and identify the demographic characteristics and possible risk factors found among these patients.&#x0D; &#x0D; Methods&#x0D; Study Design: Retrospective case series&#x0D; Setting: Private tertiary hospital&#x0D; Subjects: Records of 2,375 patients who underwent laryngeal video endoscopy and stroboscopy from 2012-2014 were reviewed.&#x0D; Results: There were 632 records of patients with benign vocal fold lesions, of which nodules were most common (211, 33.38%), followed by Reinke’s edema (165, 26.10%), cysts (122, 19.30%), and polyps (74, 11.70%), with hoarseness as the most common symptom (542, 85.76%). More than half (336, 53.16%) were aged 21-40 years, and almost two-thirds (469, 74.21%) were female. The most common associated factors were caffeine intake (445, 70.41%) and inadequate water intake (370, 58.54%), followed by alcohol (253, 40.03%). Smoking was only present in 146 (23.19%).&#x0D; Conclusion: Baseline evidence on the prevalence of benign vocal fold lesions in this institution as well as baseline data on the common characteristics and associated factors seen in the sample population may assist us in current practices and guide future studies directed toward treatment and prevention.&#x0D; &#x0D; Keywords: Vocal Cord; Stroboscopy; Vocal Cord Nodules; Benign Vocal Cord; Stroboscopy/Benign; Stroboscopy/Nodules&#x0D;

Real-time dual visualization of two different modalities for the evaluation of vocal fold vibration - Laryngeal videoendoscopy and 2D scanning videokymography: Preliminary report.

Currently, various tools have been introduced for the assessment of vocal fold vibration: laryngeal videolaryngoscopy (LV), videokymography (VKG), high speed videoendoscopy (HSV), digital videokymography (DKG), and 2D scanning videokymography (2D VKG). Among these, the authors have recently designed a dual modality examination system using LV and 2D VKG for more detailed information regarding the vibrations of the vocal folds. The clinical availability of this hybrid system offers medical imaging departments a range of potential advantages in the evaluation of vocal fold vibration. The obvious benefit of simultaneous acquisition is the improved integration of information that allows not only optimal anatomic localization, but also physical movement patterns. Other advantages include the lessened inconvenience to patients due to no longer requiring repeated examinations and shortening the examination time, and increased profitability. The purpose of study was to identify the efficacy of real-time dual examination of two different modalities for the evaluation of vocal fold vibration in human subjects and vocal fold vibration simulator. One vocally healthy subject and three patients with vocal fold nodules, a vocal cyst, and vocal fold paralysis took part in this study. The vibratory patterns of the vocal folds were visualized using simultaneous real-time examination of two different modalities. Additionally, qualitative and quantitative analyses of the dual LV and 2D VKG images were performed. Real-time dual examination using a two modality system provided high definition images of the vibratory movements of the vocal folds. By assessing the obtained images, we confirmed that the dual modality examination method was useful in the evaluation of pathologic vibratory patterns, even in non-periodic phonation. The present system might improve the understanding of the processes of vocal fold vibration and make a contribution to pathologic voice research, as well as clinical practice.

Associations between laryngeal and cough dysfunction in motor neuron disease with bulbar involvement.

True vocal fold (TVF) dysfunction may lead to cough ineffectiveness. In individuals with motor neuron disease (MND), cough impairment in the context of dysphagia increases risk for aspiration and respiratory failure. This study characterizes differences and associations between TVF kinematics and airflow during cough in individuals with bulbar MND. Sequential glottal angles associated with TVF movements during volitional cough were analyzed from laryngeal video endoscopy examinations of adults with bulbar MND (n = 12) and healthy controls (n = 12) and compared with simultaneously collected cough-related airflow measures. Significant group differences were observed with airflow and TVF measures: volume acceleration (p ≤ 0.001) and post-compression abduction TVF angle average velocity (p = 0.002) were lower and expiratory phase rise time (p = 0.001) was higher in the MND group. Reductions in maximum TVF angle during post-compression abduction in the MND group approached significance (p = 0.09). All subjects demonstrated complete TVF and supraglottic closure during the compression phase of cough, except for incomplete supraglottic closure in 2/12 MND participants. A strong positive relationship between post-compression maximum TVF abduction angle and peak expiratory cough flow was observed in the MND group, though it was not statistically significant (r = 0.55; p = 0.098). Reductions in the speed and extent of TVF abduction are seen during the expulsion phase of cough in individuals with MND. This may contribute to cough impairment and morbidity.

On transition between voice registers: Data from high-speed laryngeal videoendoscopy

How specific aspects of vocal fold vibration alter voice register has long been a subject of interest. Transitions between vocal registers are often studied using dynamic vocal fold models and electroglottographic signals. Although laryngeal high-speed videoendoscopy has also been applied to study steady-state voice registers, there has been little such investigation on the transitions between registers. In this study, we examined voice register transitions using phonations in which vocal qualities varied continuously in fundamental frequency (F0), loudness, or voice quality (from breathy to pressed). Glottal area measures {open quotient and alternating-current to open quotient ratios [Chen et al., JASA 133, 1656–1666 (2013)]} and acoustic measures (F0, energy, and H1-H2) were studied using simultaneously collected laryngeal high-speed videoendoscopy and audio recordings from 15 subjects. The video recordings were collected transorally at 10 000 frames per second. Glottal area waveforms were extracted using GlotAnTools (Version 5) [Erlangen, Germany], and acoustic measures were gathered using VoiceSauce software and analysis-by-synthesis. Glottal area measures were compared to acoustic measures and their relationships to voice register transitions were explored.

A physiologically and perceptually motivated voice source model

Many glottal source models have been proposed, but none has been systematically validated perceptually. Our previous work showed that model fitting of the negative peak of the flow derivative is the most important predictor of perceptual similarity to the target voice. In this study, a new voice source model motivated by high-speed laryngeal videoendoscopy is proposed to capture perceptually-important source shape aspects. Six voice source models (the proposed model, two previous models developed at UCLA, as well as the Fujisaki-Ljungkvist, Liljencrants-Fant, and Rosenberg models) were fitted to 40 natural voices obtained by inverse filtering and analysis-by-synthesis (AbS). We generated synthetic copies of the voices using each modeled source pulse, with all other parameters held constant, and then conducted a visual sort-and-rate task in which listeners assessed the extent of perceived match between the original natural voice samples and each copy. Model fitting results showed that the proposed model provides a more accurate fitting to the AbS-derived source than the other models. Perceptual experiments showed that the proposed model provides a close match to the original natural voice. Perceptual studies examining the extent to which each model matches the target tokens will also be reported. [Work supported by NSF grant IIS-1018863 and NIH/NIDCD grant DC01797.]

Endoscopic Assessment of Vocal Fold Movements during Cough

Little is known about the function of the true vocal folds (TVFs) during cough. The objective of this study was to determine the reliability of measuring TVF movements during cough and to obtain preliminary normative data for these measures. Sequential glottal angles associated with TVF adduction and abduction across the phases of cough were analyzed from laryngeal videoendoscopy records of 38 young healthy individuals. The intraobserver and interobserver reliability of 3 experienced measurers was high (intraclass correlation of at least 0.97) for measuring sequential and maximum glottal angles. The TVF abduction velocity during expulsion was significantly higher than the precompression adduction velocity (p = 0.002), but there were no significant differences in maximum angle. No statistically significant differences were seen in maximum TVF angle and velocity when they were compared between the sexes and between the levels of cough strength. True vocal fold closure following expulsion occurred in 42% of soft coughs and in 57% of moderate to hard coughs. The TVF abduction angles during cough can be reliably measured from laryngeal videoendoscopy in young healthy individuals. The TVF movements are faster for expulsion abduction than for precompression adduction, but the extents of abduction are similar. To validly determine the cough phase duration, simultaneous measures of airflow are needed.

A Comparison of Laryngoplasty and Modified Partial Arytenoidectomy as Treatments for Laryngeal Hemiplegia in Exercising Horses

To compare upper airway mechanics, arterial blood gases, and tracheal contamination in horses with induced left laryngeal hemiplegia (recurrent laryngeal neuropathy [RLN]) treated by laryngoplasty/vocal cordectomy (LPVC) or modified partial arytenoidectomy (MPA). Repeated measures under the following conditions: Control, RLN, LPVC, and MPA. Six horses. Two trials were conducted under all conditions at 80% and 100% of maximal heart rate (HR(max)). In Trial 1, arterial blood gases, tracheal and pharyngeal pressures, and laryngeal videoendoscopy were recorded. In Trial 2, upper airway pressure and airflow were determined. Tracheobronchial aspirates were performed after exercise to quantify airway contamination. Compared with control, RLN significantly increased inspiratory impedance and worsened exercise-induced hypoxemia. At 80% HR(max), LPVC restored most variables to control values. At 100% HR(max), LPVC improved all variables, but did not restore minute volume, arterial pH, and PaCO(2). At 80% HR(max), MPA restored all variables except bicarbonate to control values. At 100% HR(max), MPA improved all variables, but did not statistically restore minute ventilation or bicarbonate level. Only minor differences were noted between LPVC and MPA. Both resulted in equivalent tracheal contamination. Airway mechanics and arterial blood gas values were not restored to normal after either LPVC or MPA in horses exercising at HR(max). This does not affect ventilation at sub-maximal exercise, but has clinical implications at HR(max). Both procedures diminish normal laryngeal protective mechanisms. At sub-maximal exercise intensities both LPVC and MPA restore airway ventilation to normal. At maximal exercise the superiority of LPVC over MPA is slight.

Laryngeal sequelae following prolonged intubation: A prospective study

Background: Laryngeal injuries following intubation have a reported incidence from 63 to 94% and permanent sequelae are reported to be about 10 to 22% in the world literature. While several studies assessing the laryngeal complications are available in Caucasoid populations, minimal data is available in the literature regarding south Indian population. Aim: The aim of this study was to evaluate laryngeal lesions in patients after prolonged intubation (>24 h), to correlate these lesions with the variables involved in the process of intubation and to determine the risk factors. Materials and Methods: This is a prospective study for 1 year of patients who were intubated for more than 24 h in our critical care unit. Patients underwent laryngeal video endoscopy on the day of extubation and after 3 weeks by an ENT surgeon who was blinded to the intubation variables. Result: About 51 patients were included in the study. Laryngeal abnormalities were seen in 41 patients on the day of extubation. At the end of third week after extubation, only 10 patients had abnormal laryngeal findings. A multivariate stepwise regression model showed that bigger tube size ( P =0.02), longer duration of intubation ( P =0.01) and emergency intubation ( P = 0.02) was associated with higher incidence of laryngeal complications on the day of extubation. At the end of third week laryngeal findings were influenced only by the duration of intubation ( P =0.001). Conclusion: Laryngeal sequelae after extubation is directly associated with duration of intubation.