CMSD 5050 Glen Nowell Lecture 10 Phonatory Disorders Nov 8 2024 PDF

PHONATORY DISORDERS FERRAND, CHAPTER 5 Presentation design by Michael Kiefte PhD, presented by Glen Nowell, MSc., SLP-Reg November 8, 2024  How do we figure out abnormal voice? 01 02 03 Figure out how Gather Com...

PHONATORY DISORDERS FERRAND, CHAPTER 5 Presentation design by Michael Kiefte PhD, presented by Glen Nowell, MSc., SLP-Reg November 8, 2024  How do we figure out abnormal voice? 01 02 03 Figure out how Gather Compare to measure normative data clients to voice normative data What can we measure ◦ How vocal folds move (visual, conductivity, aerodynamics) ◦ How stable is a sustained sound ◦ How long you can hold a sound ◦ Vocal range ◦ Intensity range ◦ Signal to noise ratio ◦ Changes to any of these parameters over the lifespan How do we analyze VF movement? ◦ Direct laryngoscopy: during surgery ◦ Indirect laryngoscopy: laryngeal mirror, rigid Hopkins rod, flexible Nasopharyngeal Laryngoscopy (NPL)). ◦ Bright light, stroboscopy, videokymography ◦ Inferential measures (aerodynamic measures, electroglottography) ◦ Videokymography one horizontal line is selected from high-speed digital image and individual lines are stacked to make a picture Methods of investigation of laryngeal physiology three problems: 1 location 2 illumination 3 rapidity of vibration indirect laryngoscopy mirror on handle placed in throat. Laryngeal Visualization direct laryngoscopy laryngoscope inserted mouth, a tube that instruments can pass through during surgery Rigid scope (eg Hopkins rod) passes through the mouth. Angle often 70 degrees but can be 0 to 90, depending on scope. Nasopharyngolaryngoscopy(NPL) (scope in the nose) Laryngeal parameters for stroboscopy ◦ Vocal fold edge (smoothness or irregularity) ◦ Glottic closure (degree and pattern) ◦ Extent of opening (how wide can they abduct—any difficulty?) ◦ Phase closure (should be half and half open vs closed) ◦ Presence/extent of mucosal wave (inferior to superior and lateral spread of the margins of the VF) ◦ Non-vibrating portion (adynamic segment) ◦ Amplitude of vibration ◦ Phase symmetry (does one side chase the other?) ◦ Periodicity/regularity ◦ Supraglottic activity (AP squeeze, medial compression of false VF) ◦ Vertical phase symmetry/difference Videokymography Value of examination of Vocal folds ◦ Identify anatomical and physiological causes of dysphonia (paralysis, lesions, etc.) ◦ Look at condition of mucosa (edema, erythema, dryness, stiffness, injury) ◦ Look for signs of reflux, which can affect the voice reflux can irrate especially the back of the vocal folds Simple voice measures ◦ Maximum phonation time: how long you can hold a note. Measured in seconds. ◦ Pitch range (lowest pitch and highest pitch, measured in Hertz (Hz). ◦ Intensity range (softest and loudest sound), measured in decibels (dB). More sophisticated measures ◦ Signal to noise ratio ◦ Perturbation measures (Jitter and Shimmer) “good enough pitch” ◦ Cepstral Peak Prominence severely bad/ normal voice ◦ Spectrogram (visual data, useful for speech as well as voice) ◦ Long term average spectrum (visual data showing relative intensity of harmonics Changes in the Phonatory/Laryngeal system over time ◦ Vocal fold composition: Infant VF are 3 mm in length, mostly membranous portion. MFo: 400 to 600 Hz. Vocal fold structure is simple, without 3-layer structure starting until age 4. ◦ Gradual lengthening of VF leads to 230 Hz MFo in pre-teen years. ◦ Androgenic or Estrogenic Adolescence: Androgenic puberty leads to double or triple the AP dimension and weight, as well as relatively longer VF (Androgenic VF lengthen 63% compared to estrogenic VF 34%). ◦ Androgenic pitch drops an octave (12 semitones) ◦ Estrogenic drops 2.5 semitones Changes in the Phonatory/Laryngeal system over time ◦ Adulthood: ossification and deterioration of laryngeal structures can be noted in 30-40’s. ◦ Adult pitch (androgenic) ~107-123 Hz, and (estrogenic) ~191-224 Hz. ◦ Presbylaryngis: age-related changes to larynx (muscle mass loss, collagen loss, thinning of VF) ◦ Presbyphonia: age-related changes to the voice (breathiness, hoarseness, asthenia). ◦ Pitch convergence in geriatric years. ◦ Androgenic: atrophy and possibly compensatory increase of VF tension. 132-146 Hz (70-90 Yrs) ◦ Estrogenic: post-menopausal edema 160-186 Hz. How do we define voice quality ◦ Singing: chest or head register ◦ Descriptors: clear, breathy, hoarse, harsh… ◦ Context and focus of how we analyze voice affects what we call it and how we try to measure it! ◦ Super complex, too many variables to have one universally accepted way of describing voice. Timbre change in glottal tone; sound filtration ◦ Oversimplified, the sound of the tube filtering the voice is unique to each person. ◦ Our throats are as different as our faces ◦ Our throats are as similar to family as are faces—mistaken identity by phone. ◦ Trumpet vs Trombone. Same pitch, different tube length and diameter. You can tell them apart by the filtering of harmonics through the tube. Vocal fold vibration affects voice quality ◦ Interaction of adduction and Psub influence vocal fold vibration. ◦ Hyper-adduction: too much adduction, too much medial compression. PRESSED, STRAINED ◦ Hypo-adduction: not enough adduction, not enough compression: BREATHY, NOISY from turbulent air flow. Normal vocal parameters (according to Zemlin) ◦ Range ◦ Pediatric: 2 octaves ◦ Adults 2.5 to 3 octaves ◦ Speaking Fo ◦ Pediatrics: 300 Hz ◦ Adult cis females: 200 Hz ◦ Adult cis males: 100 Hz ◦ Maximum Phonation Time ◦ Children: 10 seconds ◦ Adults: 15-25 seconds ◦ Min-max intensity at various Fo levels ◦ Ability to vary intensity by 20-30 dB in mid-range ◦ Periodicity of vibration ◦ Jitter less than 1% ◦ Noise generated by turbulent airflow ◦ Normal amount of noise in signal Registers ◦ Pulse: very low, creaky poppy sound. Glottal Fry or vocal fry or creaky voice. ◦ Modal: most common register for normal conversational speech. “Heavy Mechanism” ◦ Loft/Falsetto register: higher register. “Light Mechanism” ◦ Overlap between registers. ◦ Pattern of VF vibration is different for each register. ◦ Often audible change between registers (pitch break, crack, yodel, flip). ◦ Jitter and shimmer (perturbation) increases at point of register transition. Pulse ◦ Closed phase 90%, open phase 10%. ◦ May be more than one partial approximation and separation of VF before completely adducting. “Multiphasic closure.” ◦ After each partial closure, a burst of acoustic energy is released and then ceases, leaving a Temporal gap. This Temporal gap is audible as creaky, popping sound. ◦ At the bottom end of modal register, often used at the end of sentences. ◦ SLPs treat it when it is chronic and interfering with client’s communicative functioning. Modal Register **main register for speaking; periodic; largest dynamic range ◦ Full participation of cover and body during vibration, evident mucosal wave. ◦ Biggest dynamic range in this register (loudest, softest). ◦ VF are shorter, cover is slack ◦ EGG shows brief open phase followed by a longer closing phase and a short closed phase. Loft/Falsetto ◦ Higher frequency range. ◦ Cricothyroid dominant, creates longitudinal stretch/tension ◦ VF long and stiff, edges thin and sharp, leading to a thinner and less complete sound ◦ Reduced adduction, leading to reduced intensity, somewhat breathy voice quality ◦ Higher Hz means harmonics are further apart, so sound is less rich sounding. ◦ Steeper spectral slope. Supplimentary diagnostic & research techniques (myo is muscle) ◦ Electromyography (EMG): ◦ hooked-wire electrodes implanted directly into muscle. Minimizes contaminating effects of adjacent muscles. Only indicates whether muscle is active. Using different registers in speaking and singing ◦ Lots of overlap between modal and falsetto. ◦ Abrupt change is used in some types of singing (e.g., yodelling, country, folk, soul, gospel). ◦ Classical Western style singing avoids obvious register transitions—train to make it imperceptible. Glottal spectrum ◦ Generates sound ONLY when not active in biological functions ◦ Harmonics at integer multiples of Fo ◦ Can increase Fo at will, changing harmonic spacing ◦ Can increase intensity ◦ Spectral slope~ -12dB/8ve Real-Time Spectrogram (narrow band) formant: cluster of harmonic energy that increases intensity The Value of Real-Time Spectrograms ◦ The spectrogram has been well established in the professional literature as the most comprehensive display modality for clearly revealing speech characteristics. A spectrogram shows many aspects of voice and articulation simultaneously.” ◦ www.kayelemetrics.com ◦ Gives a visual representation of the voice ◦ Illustrates Frequency and Amplitude over time. ◦ Fundamental Frequency (sung pitch) on the bottom, harmonics appear stacked above. ◦ Amplitude (relative intensity) is shown by harmonics darkening (greyscale) as they become more intense. ◦ Noise is seen between the harmonic bands ◦ Spectrogram can show relative intensity of upper versus lower harmonics (chiaroscuro) and consistency of vibrato. ◦ Breathiness, hard glottal attacks can be seen on Spectrogram. ◦ Stability and maintenance of Singer’s Formant can be seen. ◦ Rate and evenness of vibrato are visible. ◦ Visual learners have additional resource to help technique and ear training. Acoustic Analysis ◦ average Fundamental Frequency. Does it fit into predicted range based on age and gender? ◦ Frequency variability (vFo or Fo SD) is the spoken melody. Pitch sigma is semitone (ST) equivalent. 2 to 4 semitones is typical. Can also measure highest and lowest Fo and come up with the speaking range, either in Hz or ST. ◦ Frequency range used in speech decreases as we age. Tends to be greater in cis females than cis males (sociocultural?) ◦ Maximum phonational frequency range: low to high (not including pulse but including falsetto). Can be 3 or 4 octaves in young adults. May decrease with age. Voice Range Profile (Phonetogram) ◦ Envelope of voice intensity and frequency range. ◦ Maps out all frequencies at max and minimum intensity ◦ Looks like a cigar/oval, tilting slightly upward. ◦ Lowest frequencies aren’t as loud (and just a few dB range) ◦ Highest frequencies aren’t as soft (and just a few dB range) ◦ Mid-range frequencies have the biggest dynamic range (20-30 dB)  Acoustic Analysis ◦ Voice amplitude: more medial compression builds up more Psub, bigger release and sharper recoil and sharper adduction after the release gives a bigger soundwave. ◦ Normal average SPL (sound pressure level) is 60-80 dB in conversation for adults. ◦ Some clients complain of being too soft or too loud. ◦ Amplitude variability: standard deviation in dB SPL. ◦ Norm is 10 dB. More excitement increases this. Used for stress/emphasis. Waveform Perturbation measures ◦ These indicate degree of periodicity of vocal fold vibration, measured in cycle- to-cycle differences of VF vibration ◦ Frequency perturbation: Jitter ◦ Amplitude perturbation: Shimmer ◦ Jitter and Shimmer are easy to measure, commonly assessed. NOT as useful as we’d like them to be… The Problem with Perturbation measures ◦ These only work for quasi-periodic sounds! The voice has to be good enough to have a measurable frequency, before you can analyze a standard deviation from it. ◦ Diplophonia, moderate or severe dysphonia or aphonia: no perturbation measures possible. ◦ Different programs use different algorhythms to calculate perturbation. ◦ Intensity influences perturbation ◦ Quality of recording influences perturbation (mic, background noise, computer hardware, software) ◦ Poor correlation to type of voice disorder Noise measures ◦ Harmonics-to-noise ratio (HNR) ◦ Noise-to-harmonics ratio (NHR) ◦ Low HNR is little signal and much noise. We want high HNR. ◦ Low NHR is low noise and strong signal. We want low NHR. ◦ Noise is irregular vibration, turbulence from too much air escaping, etc. ◦ Look at Ferrand p. 163 for the radically different normative measures for HNR. ◦ Conversion using formula. HNR = 20 x LOG (1/NHR)… formula not on the exam. Open & speed quotients open quotient (OQ) proportion of time glottis is open during single cycle closed quotient (CQ) proportion of time glottis is closed during single cycle CQ = 1 − OQ speed quotient (SQ) ratio of abduction to adduction sometimes glottis never closes.  Electroglottography laryngography/electroglottography measures changes to conductance to current across thyroid produces Lx waveform strength of electrical signal from WEAK to STRONG Electroglottography  Why use Acoustic and Visual measures of Phonatory Function? ◦ Baseline data to compare with progress and ultimate outcome—efficacy and patient satisfaction ◦ Feedback as a therapeutic tool ◦ Can detect subtle things the ear cannot detect ◦ Can be used to screen people for potential problems as yet undetected ◦ Validates perceptual judgements with normative data as comparison Abnormal voice qualities ◦ DYSPHONIA: voice that sounds deviant re: quality, pitch and/or loudness ◦ Causes can vary, hence ENT examination before voice therapy ◦ Subjective terms are not measurable, don’t reflect underlying VF vibratory pattern ◦ Hoarse, harsh, strident, pressed, nasally, tinny, breathy, unpleasant, angry, too low, too high, “like Satan had a baby with Donald Duck,” etc. ◦ Most currently accepted are breathy and rough/hoarse ◦ Breathy: audible air escaping during phonation ◦ Rough: raspy noise in voice, giving the perception of a lower pitch ◦ Hoarse: rough plus breathy Breathy voice ◦ Loss of harmonic strength ◦ More high frequency aperiodic energy (noise) above 5 kHz and less acoustic energy (harmonics) in 2-5 kHz range. ◦ Less efficient, air loss shortens MPT, decreases dB ◦ In some languages, aspirated and unaspirated voice distinguishes between phonemes (E.g., Zulu differentiates /k/ and /kh/ voiceless stops) ◦ Common feature of organic (e.g., vocal fold paralysis), functional voice disorders. Breathy voice ◦ High NHR (speckles to stripe ratio, subjectively) Rough /hoarse voice ◦ spectral noise from flow of turbulent noise through the glottis and how periodic VF vibration is. ◦ Can be related to functional voice disorders, lesions, laryngitis or even CA. What can we expect from measuring different populations? ◦ Parkinson’s ◦ Adductor Spasmodic Dysphonia ◦ Benign vocal fold lesions ◦ Unilateral vocal fold paralysis/paresis Acoustics of rough/hoarse voice ◦ Inflammation/irritation of VF is often asymmetrical. ◦ Inflammation can alter cover from vibrating freely, affecting mucosal wave. ◦ Aperiodicity is often in the lower Frequencies, below 1000 Hz. Categories of Voice Disorders ◦ Neurogenic ◦ Psychogenic ◦ Functional ◦ Organic/Traumatic Neurogenic ◦ Central or peripheral ◦ May be progressive ◦ This category crosses many of our roles as SLPs (speech, voice, dysphagia, cognition). ◦ Sometimes we help with the identification of a disorder through its initial presentation (e.g., ALS). Classification of Neurogenic voice disorders based on phonatory dysfunction (C&C p.124) ◦ 1. Adduction or abduction problems ◦ Hypoadduction (MG, Parkinson’s, peripheral nerve paresis/paralysis) ◦ Hyperadduction (ADSD, Huntington’s, Pseudobulbar palsy) ◦ Malabduction (Shy-Drager/Parkinson’s Plus {progressive abductor paralysis}) (error on the chart). MS can also cause bilateral abductor paralysis Classification of Neurogenic voice disorders ◦ 2: Phonatory Stability ◦ A) short-term (jitter and shimmer): most neurological disorders ◦ B) long-term (tremor): Essential tremor, Parkinsonism, ALS ◦ 3: Phonatory Incoordination: ABSD ◦ 4: Mixed Disorders: Cerebellar ataxia, MS (and ALS) Psychogenic voice disorders ◦ Voice reflects emotional state ◦ Psychosocial issues can underlie or contribute to voice disorders ◦ In aphonia with psychogenic etiology, may be a trauma or stressor antecedent event. ◦ “I woke up and my voice was just gone!” Organic/Traumatic voice disorders ◦ Phonatory or non-phonatory ◦ Phonatory are typically from misuse or abuse ◦ Non-phonatory may be traumatic, medical or iatrogenic Benign Midcord Lesions (Consistent with Nodules) Papilloma (associated with HPV) Let’s see if the EGG is cooperating… Data EGG can gather: Fo, HNR, jitter, shimmer Let’s see what the perturbation measures are ◦ Normal voice ◦ Breathy voice ◦ Diplophonic voice ◦ Predict HNR, Pitch, ability to gather perturbation measures References ◦ Speech Science: An Integrated Approach to Theory and Clinical Practice (4th Ed). Ferrand, Carol T. Pearson, 2018. ◦ Classification of Voice Modality Using Electroglottogram Waveforms. September 2016 ◦ DOI:10.21437/Interspeech.2016-1194 Conference: Interspeech 2016 ◦ Authors: Borsky, Mehta, Gudjohnsen, Guðnason