Quantitative analysis of laryngeal EMG in normal subjects

Prediction of Voice Fundamental Frequency and Intensity from Surface Electromyographic Signals of the Face and Neck

Silent speech interfaces (SSIs) enable speech recognition and synthesis in the absence of an acoustic signal. Yet, the archetypal SSI fails to convey the expressive attributes of prosody such as pitch and loudness, leading to lexical ambiguities. The aim of this study was to determine the efficacy of using surface electromyography (sEMG) as an approach for predicting continuous acoustic estimates of prosody. Ten participants performed a series of vocal tasks including sustained vowels, phrases, and monologues while acoustic data was recorded simultaneously with sEMG activity from muscles of the face and neck. A battery of time-, frequency-, and cepstral-domain features extracted from the sEMG signals were used to train deep regression neural networks to predict fundamental frequency and intensity contours from the acoustic signals. We achieved an average accuracy of 0.01 ST and precision of 0.56 ST for the estimation of fundamental frequency, and an average accuracy of 0.21 dB SPL and precision of 3.25 dB SPL for the estimation of intensity. This work highlights the importance of using sEMG as an alternative means of detecting prosody and shows promise for improving SSIs in future development.

Potential Biophysiological Mechanisms Underlying Vocal Demands and Vocal Fatigue

Patients with complaint of vocal fatigue have perceptual, acoustic, and aerodynamic outcomes that are heterogeneous in nature. One reason may be due to different underlying biophysiological mechanisms that lead to these heterogeneous clinical presentations. Five potential mechanisms are proposed: neuromuscular, metabolic, vocal tissue, afferent, and central neural. Analytical frameworks and study designs to study these mechanisms are also addressed. A better understanding of biophysiological mechanisms of vocal fatigue can improve precision of therapeutic approaches. It can also help shift management from symptom-based to etiology-focused approaches for vocal fatigue.

Microstimulation in Different Parts of the Periaqueductal Gray Generates Different Types of Vocalizations in the Cat

In the cat four different types of vocalization, mews, howls, cries, and hisses were generated by microstimulation in different parts of the periaqueductal gray (PAG). While mews imply positive vocal expressions, howls, hisses, and cries represent negative vocal expressions. In the intermediate PAG, mews were generated in the lateral column, howls, and hisses in the ventrolateral column. Cries were generated in two other regions, the lateral column of the rostral PAG and the ventrolateral column of the caudal PAG. In order to define the specific motor patterns of the mews, howls, and cries, the following muscles were recorded during these vocalizations; larynx (cricothyroid, thyroarytenoid, and posterior cricoarytenoid), tongue (genioglossus), jaw (digastric), and respiration muscles (diaphragm, internal intercostal, external, and internal abdominal oblique). During these mews, howls, and cries we analyzed the frequency, intensity, activation cascades power density, turns, and amplitude analysis of the electromyograms (EMGs). It appeared that each type of vocalization consists of a specific circumscribed motor coordination. The nucleus retroambiguus (NRA) in the caudal medulla is known to serve as the final premotor interneuronal output system for vocalization. Although neurochemical microstimulation in the NRA itself also generated vocalizations, they only consisted of guttural sounds, the EMGs of which involved only small parts of the EMGs of the mews, howls, and cries generated by neurochemical stimulation in the PAG. These results demonstrate that positive and negative vocalizations are generated in different parts of the PAG. These parts have access to different groups of premotoneurons in the NRA, that, in turn, have access to different groups of motoneurons in the brainstem and spinal cord, resulting in different vocalizations. The findings would serve a valuable model for diagnostic assessment of voice disorders in humans.

Objectivation of laryngeal electromyography (LEMG) data: turn number vs. qualitative analysis

Purpose

This paper describes a first attempt to quantify LEMG data based on turn number calculation. The results obtained for both healthy and ailing thyroarytenoid (TA) muscles of patients with unilateral vocal fold immobility (UVFI) were compared with the respective qualitative evaluation concerning volitional activity to determine whether the two types of analyses deliver similar results.

Methods

LEMG data obtained from 44 adults with UVFI were considered for the study. Semiquantitative evaluation of TA volitional activity and turn number were assessed for the ailing and the healthy TA and the difference in percentage was calculated. Paired data were compared with the Wilcoxon signed-rank test. The volitional activity assessment and the turn number evaluation were compared with the Kruskal–Wallis test, and their relationship was tested with the Kendall rank correlation.

Results

Datasets of 27 patients were considered compatible with turns/s calculation. The results showed that complete paralysis correlated with no turns; single fiber volitional activity with 62–208 turns/s, strongly decreased volitional activity with 198–501 turns/s; and dense volitional activity with 441–1234 turns/s. On the ailing VF only, the Kruskal–Wallis test showed a statistically significant difference (p = 0.0001), and the Kendall rank correlation a positive relationship (r = 0.853,p ≤ 0.0001) between the volitional activity rating and the turn number assessment.

Conclusions

Our preliminary results showed that turn number evaluation is an effective tool to confirm LEMG qualitative analysis, and that, in combination with laryngostroboscopy and voice assessment, can help improving the accuracy of the diagnosis and prognosis and the effectiveness of the chosen therapy.

Motor organization of positive and negative emotional vocalization in the cat midbrain periaqueductal gray

Neurochemical microstimulation in different parts of the midbrain periaqueductal gray (PAG) in the cat generates four different types of vocalization, mews, howls, cries, and hisses. Mews signify positive vocal expression, whereas howls, hisses, and cries signify negative vocal communications. Mews were generated in the lateral column of the intermediate PAG and howls and hisses in the ventrolateral column of the intermediate PAG. Cries were generated in two regions, the lateral column of the rostral PAG and the ventrolateral column of the caudal PAG. To define the specific motor patterns belonging to mews, howls, and cries, the following muscles were recorded during these vocalizations: larynx (cricothyroid, thyroarytenoid, and posterior cricoarytenoid), tongue (genioglossus), jaw (digastric), and respiration (diaphragm, internal intercostal, external abdominal oblique, and internal abdominal oblique) muscles. Furthermore, the frequency, intensity, activation cascades, and turns and amplitude analyses of the electromyograms (EMGs) during these vocalizations were analyzed. The results show that each type of vocalization consists of a specific, circumscribed motor coordination. The nucleus retroambiguus (NRA) in the caudal medulla serves as the final premotor interneuronal output system for vocalization. NRA neurochemical microstimulation also generated vocalizations (guttural sounds). Analysis of the EMGs demonstrated that these vocalizations consist of only small parts of the emotional voalizations generated by neurochemical stimulation in the PAG. These results demonstrate that motor organization of positive and negative emotional vocal expressions are segregated in the PAG and that the PAG uses the NRA as a tool to gain access to the motoneurons generating vocalization.

Multi-Motor Unit Potential Analysis and Other Quantitative Techniques for Laryngeal Electromyogram

Reference values for motor unit potentials of laryngeal muscles, the thyroarytenoid (TA) and cricothyroid (CT) muscles, have been collected using multi-motor unit potential quantitative electromyogram analysis technique with concentric quantitative electromyogram-needle electrode type in 40 healthy volunteers. The method is well tolerated, easy to perform, and examination of one muscle takes usually around 5 minutes. The mean motor unit potential amplitude of both muscles was significantly larger in men than in women. The method can safely be used in clinical routine. Other quantitative electromyogram methods are also shortly reviewed.

Reliability of clinical office-based laryngeal electromyography in vocally healthy adults

OBJECTIVE

This study aimed to conduct a 3-session reliability assessment of the laryngeal electromyography (LEMG) signal in healthy participants during intensity controlled vocalization tasks. We hypothesized that vocal intensity level and testing session would affect LEMG measures.

METHODS

This prospective study used a 2-factor repeated measures design. Seven participants underwent bipolar needle LEMG of the right thyroarytenoid muscle. Data were collected over 3 testing sessions using vocalization tasks performed with visually guided intensity feedback targets (65 and 75 dB SPL). Root mean square amplitudes in microvolts were analyzed for within-session and between-session reliability.

RESULTS

The main effect for intensity was found to approach significance (F = 5.71, P = .054). However, intraclass correlation coefficients (ICCs) using a 2-factor mixed random effect model indicated poor to fair signal reliability between testing sessions (ICC = 0.56 at 65 dB, 0.40 at 70 dB). Intraclass correlation coefficients for within-session data indicated excellent reliability for all testing conditions (0.84-0.98).

CONCLUSION

Using a quantitative analysis protocol to inform an essentially qualitative technique, our results indicated that there was generally poor to fair reliability in the LEMG signal over testing sessions. Vocal intensity was an important variable that affected LEMG signal reliability. Standardization of LEMG protocols using vocal control parameters and quantitative analyses may help improve LEMG reliability in clinical settings.

Neural control of fundamental frequency rise and fall in Mandarin tones

The neural mechanisms used in tone rises and falls in Mandarin were investigated. Nine participants were scanned while they named one-character pictures that required rising or falling tone responses in Mandarin: the left insula and right putamen showed stronger activation between rising and falling tones; the left brainstem showed weaker activation between rising and falling tones. Connectivity analysis showed that the significant projection from the laryngeal motor cortex to the brainstem which was present in rising tones was absent in falling tones. Additionally, there was a significant difference between the connection from the insula to the laryngeal motor cortex which was negative in rising tones but positive in falling tones. These results suggest that the significant projection from the laryngeal motor cortex to the brainstem used in rising tones was not active in falling tones. The connection from the left insula to the laryngeal motor cortex that differs between rising and falling tones may control whether the rise mechanism is active or not.

Laryngeal electromyographic responses to perturbations in voice pitch auditory feedback

The present study was conducted to test the hypothesis that intrinsic laryngeal muscles are involved in producing voice fundamental frequency (F(0)) responses to perturbations in voice pitch auditory feedback. Electromyography (EMG) recordings of the cricothyroid and thyroarytenoid muscles were made with hooked-wire electrodes, while subjects sustained vowel phonations at three different voice F(0) levels (conversational, high pitch in head register, and falsetto register) and received randomized pitch shifts (±100 or ±300 cents) in their voice auditory feedback. The median latencies from stimulus onset to the peak in the EMG and voice F(0) responses were 167 and 224 ms, respectively. Among the three different F(0) levels, the falsetto register produced compensatory EMG responses that occurred prior to vocal responses and increased along with rising voice F(0) responses and decreased for falling F(0) responses. For the conversational and high voice levels, the EMG response timing was more variable than in the falsetto voice, and changes in EMG activity with relevance to the vocal responses did not follow the consistent trend observed in the falsetto condition. The data from the falsetto condition suggest that both the cricothyroid and thyroarytenoid muscles are involved in generating the compensatory vocal responses to pitch-shifted voice feedback.

Recommendations of the Neurolaryngology Study Group on laryngeal electromyography

The Neurolaryngology Study Group convened a multidisciplinary panel of experts in neuromuscular physiology, electromyography, physical medicine and rehabilitation, neurology, and laryngology to meet with interested members from the American Academy of Otolaryngology Head and Neck Surgery, the Neurolaryngology Subcommittee and the Neurolaryngology Study Group to address the use of laryngeal electromyography (LEMG) for electrodiagnosis of laryngeal disorders. The panel addressed the use of LEMG for: 1) diagnosis of vocal fold paresis, 2) best practice application of equipment and techniques for LEMG, 3) estimation of time of injury and prediction of recovery of neural injuries, 4) diagnosis of neuromuscular diseases of the laryngeal muscles, and, 5) differentiation between central nervous system and behaviorally based laryngeal disorders. The panel also addressed establishing standardized techniques and methods for future assessment of LEMG sensitivity, specificity and reliability for identification, assessment and prognosis of neurolaryngeal disorders. Previously an evidence-based review of the clinical utility of LEMG published in 2004 only found evidence supported that LEMG was possibly useful for guiding injections of botulinum toxin into the laryngeal muscles. An updated traditional/narrative literature review and expert opinions were used to direct discussion and format conclusions. In current clinical practice, LEMG is a qualitative and not a quantitative examination. Specific recommendations were made to standardize electrode types, muscles to be sampled, sampling techniques, and reporting requirements. Prospective studies are needed to determine the clinical utility of LEMG. Use of the standardized methods and reporting will support future studies correlating electro-diagnostic findings with voice and upper airway function.

Repetitive phonatory tasks for identifying vocal fold paresis

Vocal fold paresis may be present in patients with voice complaints. Identification of paresis is important so that appropriate neurolaryngologic evaluation can be ordered and the appropriate treatment can be offered. Repetitive phonatory tasks (RPTs) fatigue patients vocally and may elicit signs of subtle paresis. In this study, four laryngologists independently reviewed the RPT portions of routine fiberoptic voice examinations of 100 patients in a blinded fashion. All patients had presented with voice complaints, were suspected of having a movement disorder of the larynx, and had been referred for laryngeal electromyography (LEMG). Predictions were compared with LEMG results and with predictions made at the time of each initial evaluation. Although RPTs are useful to laryngologists, predictions based on the entire examination are more accurate. LEMG can confirm clinical suspicions or identify paresis missed on fiberoptic laryngeal examination.

Electronic analysis of intrinsic laryngeal muscles in canine sound production

This study explores the relationship between voice production and intrinsic laryngeal muscle (ILM) activities as expressed by orderly recruitment of their specific motor units. In 5 dogs, both the recurrent laryngeal nerve (RLN) and the vagus nerve (cranial nerve X) were stimulated via tripolar electrodes with stimulating frequencies (Fs) of 10 to 60 Hz and 0 to 7 mA during application of symmetric 600 Hz, 7 to 0 mA blocking currents. The fundamental frequency (Fo) and the intensity (I) of sounds generated by tracheal insufflation of humidified air were recorded while electromyograms of the cricothyroideus (CT), thyroarytenoideus (TA), and posterior cricoarytenoideus (PCA) were obtained via surface electrodes. Contractions of the CT were concurrently induced by stimulating the superior laryngeal nerve (SLN). The recruitment rates were highly specific and were affected by which nerve was stimulated. For the RLN, PCA ramping was lowest for Fs of < or =50 Hz. For Fs of 10 to 30 Hz, the recruitment rate of the TA was significantly steeper than that for the other ILMs, and the CT had the highest rate for Fs of 40 to 50 Hz. Conversely, for the vagus nerve, PCA recruitment was highest for Fs of > or =30 Hz. The average Fo was significantly higher with the RLN than with the vagus nerve. When the TA recruited faster than the CT (ie, via the RLN, but not the vagus nerve), the Fo was higher. While only CT ramping was significantly related to changes in sound intensity, there was a trend toward a decrease when PCA ramping was higher than CT ramping, as occurred when only the vagus nerve was stimulated. Stimulation of the SLN always increased Fo and loudness. We conclude that changes in Fo occur mainly through RLN-mediated CT and TA contraction. Loudness is controlled by the CT. The PCA exerts reciprocal coupling on both functions via the vagus nerve, and they are boosted across the board by the SLN. These findings may allow artificial manipulation of voice.

EMG-interference pattern analysis

The EMG interference pattern, built up of single motor unit action potentials, may be analyzed subjectively, or objectively by computer aided, quantitative methods, like counting of zero-crossings, counting of spikes, amplitude measurements, integration of the area under the curve, decomposition techniques, power spectrum analysis and turn/amplitude analysis. Since the shape of the interference pattern of healthy muscles is dependent on age, sex, force, muscle, temperature, fatigue, fitness level, recording site and surrounding tissue, electrode type, sensitivity, filters, sampling frequency and threshold level, all methods of analyzing the IP have to be standardized. Quantitative methods of analyzing the EMG interference pattern may be used for monitoring botulinum toxin therapy of dystonia and spasticity, quantifying spontaneous activity, assessment of chronic muscle pain, neuro-urological and proctological function, and diagnosing neuromuscular disorders. For diagnostic purposes, the methods favored are those that use needle electrodes and do not require measurement or monitoring of muscle force. The most well-evaluated methods are those using turn/amplitude analysis, like the cloud methods and the peak-ratio analysis. Peak-ratio analysis has the advantage that reference limits are easy to obtain and that its utility is well established and confirmed by several investigations. Overall, automatic methods of EMG interference pattern analysis are powerful tools for diagnostic and non-diagnostic purposes.

The study of laryngeal muscle activity in normal human subjects and in patients with laryngeal dystonia using multiple fine-wire electromyography

The normal human larynx performs numerous complex tasks with nearly complete reliability. These tasks require precise timing of movements that are effected by the laryngeal muscles. The most specific method to examine these muscles is by electromyography. Although many studies on laryngeal electromyography have been reported using multichannel recordings, none has provided a detailed analysis of each laryngeal muscle's role during a variety of common tasks and the spectrum of normative values. Simultaneous eight-channel, fine-wire electromyographic recordings were made in 11 human subjects. The timing patterns of the laryngeal muscles during the coordinated efforts for phonation and other common glottic functions were examined. In addition, normative values for latencies and amplitudes of response were determined. During simple phonation, a "set pattern" for the thyroarytenoid, lateral cricoarytenoid , and interarytenoid muscles was found. The thyroarytenoid and lateral cricoarytenoid muscles demonstrated a burst at onset preceding phonation and then decreased activity, whereas the interarytenoid sustained glottic position during phonation. The coordination of the laryngeal muscles was similarly determined for connected speech, respiration, Valsalva maneuver, cough, throat-clear task, and swallow. These patterns of response, the latencies for activities, and the amplitudes of response in normal subjects provided the basis to examine the abnormal laryngeal function in a group of 59 patients with four clinical varieties of laryngeal dystonia (adductor, tremor, abductor, and mixed). The findings include abnormal patterns of response, increased latencies, and increased amplitudes of recruitment in many tasks including nonphonatory tasks. Although specific distinctions were noted in each group, the responses were remarkably similar, indicating that all clinical varieties of laryngeal dystonia should be classified as mixed dystonia with a clinical preponderance for one or more types of behavior.

Differential activity of the pars recta and pars oblique in fundamental frequency control

This study was designed to determine if differences exist in pars recta and pars oblique muscle activity during speech and singing. Hooked wire electrodes were implanted in the muscle bundles under direct vision during thyroid surgery in two men and three women. It was found that the pars recta and pars oblique do not function in a similar manner across fundamental frequencies (fo's), tasks, or subjects. Large inter- and intrasubject variability was evident in the contribution of the cricothyroid bundles to fundamental frequency (fo) control. It is speculated that the effect of pars recta and pars oblique contraction may be a function of individual anatomic variations.

Laryngeal activity during swallow, phonation, and the Valsalva maneuver: an electromyographic analysis

To better understand the mechanisms of airway protection during swallow, the authors of this study performed an electromyographic (EMG) analysis on the thyroarytenoid (TA) and interarytenoid (IA) muscles during a variety of tasks. The tasks included high, low, and comfortable pitch phonation, the Valsalva maneuver, saliva swallow, and 5- and 10-mL water swallows. Raw EMG signals were analyzed to obtain root mean square data, which correspond to a relative magnitude of muscle activation. The data show that both TA and IA muscles generate a similar level of relative activation, with the greatest electrical activity observed during swallow tasks followed by the Valsalva maneuver and phonation. The duration, onset, offset, and pattern of activity during the swallowing tasks also showed close synchronization between the two muscles. These data can be used in designing therapy for voice disorders and pharyngeal dysphagia.

Long-term effects of botulinum toxin injections in spasmodic dysphonia

The purpose was to examine whether physiological changes can be found in laryngeal muscles following repeated treatment with botulinum toxin injections in spasmodic dysphonia. Seven patients whose treatment consisted of multiple unilateral thyroarytenoid injections were examined more than 6 months following their most recent botulinum toxin injection fiberoptic laryngoscopy and electromyography. Comparisons were made between injected and contralateral noninjected muscles' motor unit characteristics, muscle activation patterns, and vocal fold movement characteristics. The results demonstrated that motor unit characteristics differed between injected and noninjected muscles and that these differences were greater in patients less than 12 months since last injection. Motor unit duration differences were reduced and motor unit amplitude and numbers of turns were increased in muscles sampled over 1 year after injection. These results suggest that while the physiologic effects of botulinum toxin are reversible, the reinnervation process continues past 12 months following injection.

Quantitative analysis of EMG interference pattern in patients with laryngeal paresis

Conventional (visual) analysis of EMG was compared to a computerized quantitative method (turns/amplitude). Ten patients with a clinical history of peripheral nerve lesion were examined in order to investigate whether quantitative analysis of EMG increased the diagnostic sensitivity of peripheral nerve lesions to the laryngeal muscles. Quantitative analysis either agreed with conventional analysis or resulted in false negative diagnosis compared to conventional analysis. Thus, conventional analysis turned out to be more sensitive to neurogenic damage than the quantitative method. EMG results were compared to clinical findings.