A Machine-Learning Algorithm for the Automated Perceptual Evaluation of Dysphonia Severity

OBJECTIVES

Auditory-perceptual assessments are the gold standard for assessing voice quality. This project aims to develop a machine-learning model for measuring perceptual dysphonia severity of audio samples consistent with assessments by expert raters.

METHODS

The Perceptual Voice Qualities Database samples were used, including sustained vowel and Consensus Auditory-Perceptual Evaluation of Voice sentences, which were previously expertly rated on a 0-100 scale. The OpenSMILE (audEERING GmbH, Gilching, Germany) toolkit was used to extract acoustic (Mel-Frequency Cepstral Coefficient-based, n = 1428) and prosodic (n = 152) features, pitch onsets, and recording duration. We utilized a support vector machine and these features (n = 1582) for automated assessment of dysphonia severity. Recordings were separated into vowels (V) and sentences (S) and features were extracted separately from each. Final voice quality predictions were made by combining the features extracted from the individual components with the whole audio (WA) sample (three file sets: S, V, WA).

RESULTS

This algorithm has a high correlation (r = 0.847) with estimates of expert raters. The root mean square error was 13.36. Increasing signal complexity resulted in better estimation of dysphonia, whereby combining the features outperformed WA, S, and V sets individually.

CONCLUSION

A novel machine-learning algorithm was able to perform perceptual estimates of dysphonia severity using standardized audio samples on a 100-point scale. This was highly correlated to expert raters. This suggests that ML algorithms could offer an objective method for evaluating voice samples for dysphonia severity.

LEVEL OF EVIDENCE: 4

Cannabinoid Use in the Treatment of Laryngeal Dystonia and Vocal Tremor: A Pilot Investigation

OBJECTIVES/HYPOTHESIS

Laryngeal dystonia and vocal tremor can be debilitating conditions with suboptimal treatment options. Botulinum toxin chemodenervation is typically the first-line treatment and is considered the gold standard. However, patient response to botulinum toxin varies widely. There is anecdotal evidence for the use of cannabinoids in treating laryngeal dystonia with a scarcity of research investigating this potential treatment option. The primary objective of this study is to survey patients with laryngeal dystonia and vocal tremor to gauge how some people are using cannabinoids to treat their condition and to ascertain patient perceptions of cannabinoid effectiveness.

STUDY DESIGN

This is a cross-sectional survey study.

METHODS

An eight-question anonymous survey was distributed to people with abductor spasmodic dysphonia adductor spasmodic dysphonia, vocal tremor, muscle tension dysphonia, and mixed laryngeal dystonia via the Dysphonia International (formerly National Spasmodic Dysphonia Association) email listserv.

RESULTS

158 responses: 25 males and 133 females, (mean [range] age, 64.9 [22-95] years). 53.8% of participants had tried cannabinoids for the purposes of treating their condition at some point, with 52.9% of this subset actively using cannabis as part of their treatment. Most participants who have used cannabinoids as a treatment rank their effectiveness as somewhat effective (42.4%) or ineffective (45.9%). Participants cited a reduction in voice strain and anxiety as reasons for cannabinoid effectiveness.

CONCLUSIONS

People with laryngeal dystonia and/or vocal tremor currently use or have tried using cannabinoids as a treatment for their condition. Cannabinoids were better received as a supplementary treatment than as a stand-alone treatment.

Local Reaction Causing Airway Compromise Following Vocal Fold Injection Augmentation with Carboxymethylcellulose

OBJECTIVES

This case report aims to identify local allergic reaction as a rare complication of vocal fold injection augmentation with carboxymethylcellulose and discuss the management of airway swelling following this reaction.

INTRODUCTION

Glottis insufficiency due to true vocal fold immobility is important to manage to reduce risk of aspiration and improve voice function. Vocal fold injection augmentation with carboxymethylcellulose is considered a safe and effective treatment for glottis insufficiency, commonly due to vocal fold immobility.

METHODS

Retrospective medical records review and case report.

RESULTS

We report a unique case of an adult female with vocal fold immobility that was treated with injection laryngoplasty using carboxymethylcellulose and subsequently developed a local reaction requiring intubation and tracheostomy placement.

CONCLUSION

Otolaryngologists should be aware of this rare, but life-threatening complication and counsel patients accordingly when obtaining consent. If there are signs and symptoms of airway edema, the patient should be transferred to the ICU for airway watch, intravenous steroids, and possible intubation.

Safety of Platelet-Rich Plasma Subepithelial Infusion for Vocal Fold Scar, Sulcus, and Atrophy

OBJECTIVE

To demonstrate the safety profile of platelet-rich plasma (PRP) as an injectable therapeutic for the treatment of vocal fold scarring and atrophy.

METHODS

Preliminary report on a prospective clinical trial of patients with vocal fold scar or atrophy undergoing unilateral vocal fold subepithelial infusion with autologous PRP. Enrolled patients underwent four subepithelial injections spaced 1 month apart. Adverse events were assessed peri and post-injection at each session. Patient-reported outcomes were collected at every visit using the Voice Handicap Index-10 (VHI-10) and Vocal Fatigue Index (VFI) questionnaires.

RESULTS

Twelve patients underwent unilateral vocal fold injection with autologous PRP prepared according to Eclipse PRP® system protocol. Forty-three injections were performed using a peroral or percutaneous approach. An average of 1.57 ± 0.4 cc (range 0.6-2.0 cc) injectate was used. All patients tolerated the procedure without difficulty or peri-procedural complications. The average duration of follow-up was 3.6 ± 1.8 months. No significant inflammatory reactions or adverse events were seen to date. There was statistically significant improvement in patient-reported outcomes at the 3 month follow up (n = 9) follow-up (mean ΔVHI-10 = 10.8, p < 0.001, mean ΔVFI = 18.9, p = 0.01, t test, paired two sample for means, two-tail). All nine patients who completed the series of four injections subjectively (yes/no) reported they were satisfied with the results.

CONCLUSION

This prospective study cohort demonstrated a favorable safety profile, with no adverse events or peri-procedural complications. Subjective improvements in vocal quality and reduction in vocal fatigue need to be clinically correlated with further study.

LEVEL OF EVIDENCE

4 Laryngoscope, 133:647-653, 2023.

Cricothyroid Muscle Botulinum Toxin Chemodenervation to Treat Recalcitrant High-Pitched Functional Dysphonia in an Adult Male

INTRODUCTION

Functional dysphonia occurs in the absence of structural or neurologic laryngeal abnormalities with various manifestations including aphonia and inappropriate vocal pitch. Voice therapy is the first-line treatment, and literature on treatment options for functional dysphonia unresponsive to voice therapy is limited.

METHODS

Retrospective medical records review and report of a case.

RESULTS

We report a unique case of an adult male with recalcitrant high-pitched functional dysphonia of adult-onset that was successfully treated with a single cricothyroid muscle botulinum toxin (Botox) chemodenervation injection.

CONCLUSION

Cricothyroid muscle botulinum toxin chemodenervation can be an effective treatment for recalcitrant functional dysphonia.

Evaluation of Acoustic Analyses of Voice in Nonoptimized Conditions

Objectives This study aimed to evaluate the fidelity and accuracy of a smartphone microphone and recording environment on acoustic measurements of voice. Method A prospective cohort proof-of-concept study. Two sets of prerecorded samples (a) sustained vowels (/a/) and (b) Rainbow Passage sentence were played for recording via the internal iPhone microphone and the Blue Yeti USB microphone in two recording environments: a sound-treated booth and quiet office setting. Recordings were presented using a calibrated mannequin speaker with a fixed signal intensity (69 dBA), at a fixed distance (15 in.). Each set of recordings (iPhone-audio booth, Blue Yeti-audio booth, iPhone-office, and Blue Yeti-office), was time-windowed to ensure the same signal was evaluated for each condition. Acoustic measures of voice including fundamental frequency (f_o), jitter, shimmer, harmonic-to-noise ratio (HNR), and cepstral peak prominence (CPP), were generated using a widely used analysis program (Praat Version 6.0.50). The data gathered were compared using a repeated measures analysis of variance. Two separate data sets were used. The set of vowel samples included both pathologic (n = 10) and normal (n = 10), male (n = 5) and female (n = 15) speakers. The set of sentence stimuli ranged in perceived voice quality from normal to severely disordered with an equal number of male (n = 12) and female (n = 12) speakers evaluated. Results The vowel analyses indicated that the jitter, shimmer, HNR, and CPP were significantly different based on microphone choice and shimmer, HNR, and CPP were significantly different based on the recording environment. Analysis of sentences revealed a statistically significant impact of recording environment and microphone type on HNR and CPP. While statistically significant, the differences across the experimental conditions for a subset of the acoustic measures (viz., jitter and CPP) have shown differences that fell within their respective normative ranges. Conclusions Both microphone and recording setting resulted in significant differences across several acoustic measurements. However, a subset of the acoustic measures that were statistically significant across the recording conditions showed small overall differences that are unlikely to have clinical significance in interpretation. For these acoustic measures, the present data suggest that, although a sound-treated setting is ideal for voice sample collection, a smartphone microphone can capture acceptable recordings for acoustic signal analysis.

Robot-Automated Cartilage Contouring for Complex Ear Reconstruction: A Cadaveric Study

OBJECTIVES/HYPOTHESIS

Auricular reconstruction requiring manual contouring of costal cartilage is complex and time consuming, which could be facilitated by a robot in a fast and precise manner. This feasibility study evaluates the accuracy and speed of robotic contouring of cadaver costal cartilage.

METHODS

An augmented robot with a spherical burr was used on cadaveric rib cartilage. Using a laser scanner, each rib section was converted to a three-dimensional model for preoperative planning. A model ear was also scanned to define a carving path for each piece of cartilage. After being contoured, each specimen was compared against the preoperative plan utilizing deviation maps to analyze topographic accuracy. Contouring times of the robot were compared with 13 retrospectively reviewed cases (2006-2017) by an experienced surgeon.

RESULTS

Scanning the cartilage sections took 24.8 ± 6.8 seconds. Preoperative processing took an additional 29.9 ± 8.9 seconds for the preparation of the contouring path. Once the path was prepared, the robot contoured the specimens with a root mean square error of 0.54 mm and a mean absolute deviation of 0.40 mm. The average time to contour the specimens with the robot was 13 ± 2 minutes compared to 71 ± 6 minutes for the manual contouring in the reviewed cases.

CONCLUSIONS

The accuracy of the robotic system was high, with submillimeter deviations from the preoperative plan. The robot required <20% of the contouring time compared to the experienced surgeon. This represents a fast and accurate alternative to hand-contouring costal cartilage grafts for auricular reconstruction. Laryngoscope, 131:1002-1007, 2021.

Functional outcomes in early (T1/T2) supraglottic cancer: a systematic review

OBJECTIVES

Organ preserving surgery (OPS) and radiotherapy (RT) are both accepted treatment options for early stage supraglottic cancer (SGC). Radiation has supplanted surgery in most cases, because of the perception that surgery results in poorer functional outcomes. However, evidence suggests that OPS with a neck dissection may be associated with improved survival. Our objective was to conduct a systematic review of the literature to compare functional outcomes of OPS and RT for early SGC.

METHODS

We searched Medline, EMBASE and Cochrane Central Register of Controlled Trials to identify studies. Studies were included if they reported functional outcomes on 10 or more patients with early stage SGC treated with radiation or OPS, including open partial laryngectomy, transoral laser microsurgery (TLM) or transoral robotic surgery (TORS). Two reviewers independently screened articles for relevance using pre-determined criteria.

RESULTS

From 7720 references, we included 10 articles (n = 640 patients). 50% (n = 320) of patients were treated with surgery. Three head-to-head RT versus OPS papers were included, however different outcome measures were used for each group. Intractable aspiration management (including total laryngectomy or permanent tracheostomy) following OPS was reported in five papers representing 186 patients; the definitive intractable aspiration management rate was 2.6% (95% CI 1.0-6.8%). Four papers reported permanent G-tube rate for the surgical group (n = 198), calculating a rate of 5.3% (95% CI 2.6-10.5%), this was not reported for the RT group in any papers. One study reported quality of life. Two studies reported objective voice measures.

CONCLUSIONS

This systematic review revealed a paucity of objective measures and significant data heterogeneity, rendering the comparison of functional outcomes following OPS versus RT for early SGC limited. Future research should include objective measures of functional outcomes including laryngectomy rate, g-tube rate, tracheostomy dependence, quality of life, and voice quality measures.