🩺 Otoacoustic Emission Cochlear Function Screening Simulator

Prestin-Driven Electromotility — How a Healthy Cochlea Manufactures Sound It Never Received

Otoacoustic emissions exist only because the mammalian cochlea is not a passive receiver — it is an active, energy-consuming amplifier. Outer hair cells convert receptor potential oscillations directly into mechanical length changes at auditory rates, feeding energy back into the traveling wave on the basilar membrane. A small fraction of that mechanical energy propagates backward through the middle ear ossicular chain and radiates into the ear canal as a faint, measurable sound — the otoacoustic emission that this entire screening technology is built to detect.

• 4–5%: OHC length change (per stimulus cycle, up to 20 kHz)
• 40–60 dB: Cochlear amplification (gain over passive basilar membrane motion)
• Prestin: Motor protein (SLC26A5, OHC lateral wall membrane)
• ~12,000: OHC count (human) (vs. ~3,500 inner hair cells (IHCs))

Somatic electromotility and the physiological basis of emission generation

The organ of Corti sits atop the basilar membrane inside the cochlear duct, and its three rows of outer hair cells are the engine behind both normal hearing sensitivity and otoacoustic emissions:

Mechanotransduction and receptor potential: • Sound-driven basilar membrane motion deflects OHC stereocilia bundles • Deflection toward the tallest row opens mechanotransduction (MET) channels (TMC1/TMC2 complex) • K+/Ca2+ influx depolarizes the OHC — receptor potential tracks stimulus waveform up to several kHz

Prestin and somatic electromotility: • Prestin (SLC26A5) is a voltage-sensitive motor protein densely packed in the OHC lateral wall membrane (~10^4 copies/µm²) • Membrane depolarization → prestin conformational change → OHC body shortens • Membrane hyperpolarization → OHC body elongates • This length change (electromotility) occurs cycle-by-cycle, tracking the acoustic stimulus frequency directly — no chemical synapse or spike required • Unlike stereocilia-bundle motility (found in non-mammalian ears), somatic electromotility can operate at ultrasonic rates, matching the mammalian high-frequency hearing range

The cochlear amplifier: • OHC contraction/elongation pumps additional mechanical energy back into the traveling wave, locally amplifying basilar membrane displacement 1,000–1,000,000-fold (40–60 dB) at the stimulus place • This active process sharpens frequency tuning (higher Q, narrower tuning curves) and extends the dynamic range of hearing by ~50 dB • Amplification is nonlinear and level-dependent: compressive growth of response with increasing stimulus intensity — the same nonlinearity responsible for distortion product generation

Retrograde emission propagation: • A portion of the mechanical energy generated by OHC electromotility does not stay confined to the traveling wave — it propagates backward (base-ward) along the basilar membrane • At the stapes footplate, this reverse-traveling energy is transmitted backward through the ossicular chain (stapes→incus→malleus) in reverse of the normal forward transmission path • The tympanic membrane, driven now as an output transducer rather than an input receiver, radiates a faint acoustic pressure wave into the ear canal • This emitted sound — 15 to 30 dB below ambient noise floor, typically −5 to +15 dB SPL — is the otoacoustic emission captured by the probe microphone

Why emissions require normal OHC function: • IHC-only pathologies (auditory neuropathy spectrum disorder, ANSD) can leave OAEs intact because OHCs and the cochlear amplifier remain functional even when IHC-to-nerve signaling is disrupted • Conversely, any process that damages OHCs — noise trauma, ototoxic drugs, hereditary OHC-selective disorders (e.g. some DFNB mutations), presbycusis — reduces or abolishes OAEs well before audiometric threshold changes, because OHC amplification loss of 20–30 dB does not always cross the threshold-detection boundary at the IHC/auditory nerve level

Sealing the Canal — Sub-Microvolt Acoustic Signals Pulled Out of the Noise Floor

An otoacoustic emission arriving at the ear canal opening is an extraordinarily weak acoustic event — often 40 dB or more below the ambient noise in a quiet clinic room. The entire measurement chain, from probe fit to digital averaging algorithm, exists to recover this signal reliably and reproducibly in a few seconds per ear, which is what makes OAE testing practical as a mass screening tool rather than a research-lab-only technique.

• −5 to 15 dB SPL: Emission level (vs. ambient noise ~30–40 dB SPL)
• 260–1000+: Typical averages (stimulus sweeps per recording)
• ~10–60 sec: Test time per ear (TEOAE faster than DPOAE sweep)
• Real-time: Probe fit check (stimulus stability / leak detection)

Probe assembly, artifact rejection, and coherent signal averaging

Probe construction:

• Disposable soft foam or silicone eartip, sized to the ear canal, creates an acoustic seal preventing stimulus leakage and ambient noise ingress • Miniature receiver (speaker) delivers the calibrated click or tone stimulus; a separate low-noise microphone, positioned within 1–2mm of the receiver output, records ear canal sound pressure • In-ear calibration: stimulus level is measured and adjusted in situ for each ear canal volume (smaller neonatal canals produce higher SPL for the same driver voltage) — critical because infant canal volume is roughly one-third of adult volume

Noise floor and the averaging problem: • Ambient/physiological noise (breathing, swallowing, probe movement, environmental sound) typically sits 20–40 dB SPL in the ear canal — far above the emission itself • Time-domain coherent averaging: since the emission is phase-locked to the repeating stimulus but noise is not, averaging N repetitions increases the emission-to-noise ratio by approximately √N • 400 sweeps → theoretical SNR improvement of ~13 dB (10·log10(400)) relative to a single sweep • Practical target: 6–20+ dB SNR at frequencies of clinical interest before accepting a pass result

Artifact rejection: • Buffers exceeding an amplitude threshold (typically ±(a few) mPa-equivalent, tunable) are discarded in real time — these usually reflect probe movement, infant vocalization, or swallowing • Alternating-buffer method: sweeps are assigned alternately into buffer A and buffer B; comparing A vs. B gives both a noise estimate (difference) and signal estimate (average), and their cross-correlation becomes the reproducibility metric used in TEOAE pass/refer decisions

Stimulus stability monitoring: • The system continuously checks in-ear stimulus level and spectral shape against the calibration target • A shifted or unstable stimulus (poor probe seal, canal collapse, cerumen occlusion) triggers a "check probe fit" prompt rather than allowing a false refer or false pass • Cerumen (earwax) occlusion is among the most common causes of an artifactual "refer" result in newborn and pediatric screening, underscoring the importance of otoscopic inspection before testing

Transient-Evoked OAE — One Broadband Click, One Cochlear-Wide Snapshot

TEOAE testing uses a single brief click to excite nearly the entire length of the basilar membrane simultaneously, producing a time-domain emission waveform whose later components correspond to lower-frequency, more apical cochlear regions (because the traveling wave takes longer to reach the apex). This makes TEOAE testing fast, robust, and the dominant method used in universal newborn hearing screening programs worldwide.

• ~80 µs: Click duration (broadband, ~65–85 dB peSPL)
• ~20 ms: Analysis window (post-stimulus, onset ~2.5–20ms)
• >70%: Reproducibility pass (whole-wave A/B correlation)
• >3–6 dB: Band SNR pass (per half-octave band, 1–4 kHz)

Click stimulus design, tonotopic decomposition, and pass/refer decision logic

Why a click works as a whole-cochlea probe:

• An ~80 microsecond click contains broadband spectral energy across roughly 500Hz–6kHz in a single acoustic event • Because the cochlea is tonotopically organized (base = high frequency, apex = low frequency) and the traveling wave takes finite time to propagate, different frequency components of the emission arrive back at the ear canal at different latencies • High-frequency (basal) contributions appear early in the response window (~2.5–8ms post-stimulus); low-frequency (apical) contributions appear later (~8–20ms) • This latency structure lets a single click-evoked recording be decomposed, via time-windowing or wavelet/FFT analysis, into approximate frequency-band emission levels without needing separate tone-burst stimuli

Recording protocol: • Non-linear (derived-response) click paradigm is standard: a sequence of 4 clicks (3 at level L, 1 inverted and 3× the amplitude) is presented; linear summation cancels the linear/artifactual stimulus ringing while preserving the compressively nonlinear cochlear response — isolating true OAE energy from stimulus artifact • First ~2.5ms of the window is excluded from analysis to avoid contamination by residual stimulus ringing and middle-ear/probe artifact

Pass/refer criteria (typical clinical default settings): • Overall wave reproducibility (cross-correlation between buffer A and buffer B waveforms): must exceed 70% (some protocols use 50% for research, 70–80% for clinical screening) • Per-band SNR: response energy must exceed the noise floor by 3–6 dB in the relevant half-octave or octave bands, most commonly evaluated across 1, 1.5, 2, 3, and 4kHz • Overall response amplitude: typically must exceed roughly 3–6 dB SPL total RMS in a normal-hearing ear, though absolute-level criteria are used less often than SNR/reproducibility criteria • A "pass" requires meeting SNR and reproducibility criteria in enough bands (protocol-dependent, e.g., 3 of 5 or all bands) — a single weak band with an otherwise strong response may still be flagged for review

Limitations: • TEOAEs are typically absent or markedly reduced when hearing loss exceeds roughly 25–30 dB HL, and are insensitive to loss confined to a narrow high-frequency region above ~4–5kHz where click energy content is weaker — this is one reason DPOAE and AABR complement TEOAE in comprehensive protocols

Distortion-Product OAE — Exploiting Cochlear Nonlinearity with Two Simultaneous Tones

Where TEOAE relies on a single broadband click, DPOAE testing presents two pure tones simultaneously and exploits the cochlea's inherent mechanical nonlinearity — the same compressive nonlinearity that underlies the cochlear amplifier's wide dynamic range — to generate a distortion product at a precisely predictable frequency. Sweeping the primary tones across the audible range builds a DP-gram: a frequency-specific fingerprint of cochlear function, generally regarded as more frequency-specific than TEOAE, particularly at higher frequencies.

• ≈1.22: Primary ratio f2/f1 (optimized for maximal DP amplitude)
• 2f1 − f2: Distortion product (cubic difference tone, dominant DP)
• 65/55 dB SPL: Typical levels L1/L2 (L1 > L2 optimizes DP generation)
• 1–8 kHz (to 16 kHz): DP-gram sweep range (f2 stepped in ~1/3–1/6 octave steps)

Cubic distortion product generation and DP-gram interpretation

Where the distortion product comes from:

• Two primary tones f1 (lower) and f2 (higher, f2/f1 ≈ 1.22) are presented continuously and simultaneously through the probe • The region of the basilar membrane where the two traveling waves overlap (near the f2 place) behaves as a nonlinear mixer, because OHC transduction and electromotility are compressively nonlinear functions of stimulus level • Nonlinear mixing generates multiple intermodulation products at frequencies nf1 ± mf2; the cubic difference tone 2f1−f2 is consistently the largest and most robustly measurable in humans • Example: f1=4000Hz, f2=4880Hz (ratio 1.22) → DP = 2(4000)−4880 = 3120Hz

Stimulus level optimization: • L1=65 dB SPL, L2=55 dB SPL (L1>L2 by ~10dB) is a widely used clinical default — asymmetric levels maximize DP amplitude relative to using equal levels, based on cochlear nonlinearity growth functions mapped in the 1990s–2000s • Lower-level "sensitive" protocols (e.g., L1/L2 = 55/40 dB SPL) increase specificity for detecting mild OHC dysfunction at the cost of a somewhat higher noise floor challenge

Building and reading a DP-gram: • f2 is stepped across the frequency range of interest (commonly 1–8kHz, extended-range instruments to 16kHz) in fine steps (4–6 points per octave), each with f1 recalculated to maintain the 1.22 ratio • At each f2, DP amplitude (dB SPL) is plotted against the corresponding f2, alongside the noise floor measured at nearby non-DP frequency bins • Pass criterion: DP amplitude must exceed the noise floor by a minimum SNR (commonly ≥6 dB, sometimes frequency-dependent) at a sufficient number of test frequencies • A DP-gram with amplitude collapsing toward the noise floor at high frequencies (while low/mid frequencies remain normal) is a classic early signature of noise-induced or ototoxic OHC damage, which characteristically begins in the 3–6kHz "noise notch" or extends progressively from the extreme base (highest frequencies, shortest OHCs) inward

TEOAE vs. DPOAE — complementary, not redundant: • TEOAE: faster, single broadband stimulus, better established normative newborn database, primary tool in most UNHS programs • DPOAE: superior frequency specificity, better high-frequency (>4kHz) sensitivity, preferred for serial ototoxicity monitoring and detailed frequency-specific follow-up after a screening refer

Universal Newborn Hearing Screening — From Delivery Room to the JCIH 1-3-6 Benchmark

Congenital hearing loss affects roughly 1 to 3 infants per 1,000 live births — among the most common congenital conditions — yet before universal screening programs, the average age of identification for significant hearing loss was 2–3 years, well past the critical window for optimal speech and language development. OAE testing, paired with automated auditory brainstem response (AABR) testing, transformed newborn hearing screening into a fast, objective, low-cost bedside procedure performed on virtually every infant before hospital discharge.

• 1–3 / 1,000: Congenital HL prevalence (live births; higher in NICU graduates (~2–4%))
• ~8–10%: Initial refer rate (OAE) (single-stage OAE only)
• 2–4% → <1%: Refer after 2-stage/rescreen (OAE+AABR combined protocol)
• 1-3-6: JCIH benchmark (screen–diagnose–intervene, in months)

Two-stage OAE/AABR protocol design and the rationale for combined testing

Why a single OAE pass/refer is not the whole protocol:

• OAE testing alone has a meaningful false-positive ("refer" in a normal-hearing infant) rate, largely driven by transient middle-ear fluid or vernix/debris in the ear canal in the first 24–48 hours of life — both of which resolve without intervention • OAE also cannot detect auditory neuropathy spectrum disorder (ANSD), a disorder of IHC/auditory-nerve synchrony in which OHC function (and therefore OAEs) can be entirely normal despite significant hearing/processing impairment • AABR (automated auditory brainstem response) records synchronized neural activity from scalp electrodes in response to click or chirp stimuli, evaluating the full pathway from cochlea through the auditory nerve and brainstem — catching ANSD and neural pathway disorders that OAE alone would miss

Common two-stage/two-step protocol structures: 1. OAE first pass — if pass, infant is done (majority of infants) 2. OAE refer → immediate or same-admission AABR — if AABR passes, infant is cleared (most OAE-refer infants pass AABR, reflecting transient conductive causes) 3. AABR refer (or refer on both) → scheduled outpatient rescreen, typically within 2–4 weeks 4. Persistent refer at outpatient rescreen → referral to full diagnostic audiologic evaluation (frequency-specific ABR under sedation/natural sleep, tympanometry, case history)

Quantifying the funnel: • Single-stage OAE-only screening: ~8–10% initial refer rate • Two-stage in-hospital OAE→AABR: reduces same-admission refer rate to roughly 2–4% • After scheduled outpatient rescreen: refer rate for true diagnostic referral typically falls below 1% • Of infants ultimately referred for full diagnostic workup, a meaningful minority (device- and population-dependent, commonly cited in the range of 1 in a few dozen) are confirmed with permanent hearing loss — reflecting that "refer" is a screening flag, not a diagnosis

JCIH 1-3-6 (updated 2019 guidance also references 1-2-3 for risk-stratified pathways): • 1 month: hearing screening completed for all infants (in-hospital before discharge, or outpatient within the first month) • 3 months: audiologic diagnostic evaluation completed for anyone who did not pass screening • 6 months: enrollment in early intervention services (amplification, communication therapy, cochlear implant evaluation if indicated) for any confirmed hearing loss • Programs meeting the 1-3-6 benchmark are strongly associated with better expressive and receptive language outcomes at school age compared with later-identified children — the entire rationale for universal, not risk-factor-only, screening

Before universal newborn hearing screening became standard practice, the average age of confirmed diagnosis for congenital hearing loss was 2–3 years old — well past the first-year window when the developing auditory cortex is most plastic. Population-level studies following the rollout of UNHS programs (US, UK, and elsewhere) since the late 1990s consistently associate screening-detected, early-intervention infants with significantly better language outcomes at ages 5–8 than infants identified later through parental or caregiver concern alone.

Detecting Damage Before It Shows Up on an Audiogram — OAEs as an Early-Warning System

Because outer hair cells sit at the most exposed, most metabolically demanding position in the organ of Corti, they are consistently the first cochlear structure to fail under insult — noise trauma, ototoxic chemotherapy, or aminoglycoside antibiotics all damage OHCs well before inner hair cells or the auditory nerve are affected. OAE amplitude loss therefore precedes measurable pure-tone threshold shift, turning otoacoustic emission testing into a sensitive early-warning biomarker rather than only a pass/fail screening tool.

• ~1–3 weeks: OHC vs. audiogram lead time (DPOAE decline before threshold shift)
• 8–16 kHz DPOAE: Cisplatin ototoxicity (extended high-frequency most sensitive)
• Platinum agents, aminoglycosides: Ototoxic drug classes (cisplatin, gentamicin, tobramycin, etc.)
• 3–6 kHz notch: Noise-induced pattern (classic early NIHL DP-gram signature)

Why OHCs fail first, and how serial OAE monitoring is used clinically

The biological basis of OHC vulnerability:

• OHCs have among the highest metabolic rates of any cell type in the body, driven by the energetic cost of continuous, high-rate electromotility and ion cycling through the mechanotransduction/prestin cycle • Their position — three exposed rows sitting directly in the path of the traveling wave, with minimal structural protection compared to the single, more centrally located IHC row — leaves them mechanically and metabolically exposed to both acoustic overexposure and circulating ototoxic compounds • Ototoxic drugs preferentially accumulate in OHCs (particularly basal-turn, high-frequency-coding OHCs) via mechanotransduction-channel-mediated uptake and generate reactive oxygen species that trigger apoptotic or necrotic OHC death, typically progressing from base (high frequency) toward apex (low frequency) as cumulative dose increases • Once enough OHCs in a given cochlear region are lost, the local ~40–60 dB active amplification gain collapses — but IHCs and the auditory nerve in that same region may remain intact and still transduce sound, just without amplification, meaning the pure-tone threshold shift required to register on a standard audiogram (a 15–25+ dB change) lags behind the underlying OHC injury by a meaningful margin

Clinical monitoring protocols:

Oncology (platinum-based chemotherapy, e.g. cisplatin, carboplatin): • Baseline DPOAE/audiogram obtained before first infusion cycle • Serial DPOAE testing before each subsequent cycle, emphasizing extended high-frequency range (8–16kHz) — the earliest and most consistently affected band • A significant DP amplitude decline (commonly defined as ≥6 dB drop from baseline at 2 or more adjacent frequencies) can prompt oncology team discussion of dose modification, otoprotectant use (e.g., sodium thiosulfate in select pediatric protocols), or closer monitoring — before the loss becomes audiometrically apparent or symptomatic • ASHA/AAA ototoxicity monitoring guidelines formalize this as part of standard-of-care hearing conservation for high-risk chemotherapy regimens

Aminoglycoside antibiotics (gentamicin, tobramycin, amikacin — common in NICU, cystic fibrosis, and multidrug-resistant infection treatment): • Similar base-to-apex progressive OHC injury pattern • Serial DPOAE monitoring is especially valuable in NICU populations where behavioral/audiometric testing is not feasible

Occupational and recreational noise-induced hearing loss (NIHL) surveillance: • Classic early DP-gram signature: an amplitude notch centered around 3–6kHz, corresponding to the cochlear region of maximal mechanical stimulation from typical broadband occupational noise and the ear canal's own resonance-driven amplification near 3kHz • Hearing conservation programs increasingly explore DPOAE as a pre-symptomatic screening adjunct to annual audiometry, since DPOAE changes can precede the OSHA-defined "standard threshold shift" (10 dB average shift at 2,3,4kHz) by weeks to months in some exposed cohorts

In pediatric oncology cohorts receiving cisplatin, serial high-frequency DPOAE monitoring has been shown in multiple studies to detect a significant amplitude decline roughly 1 to 3 weeks before any corresponding change is measurable on conventional pure-tone audiometry — a window during which oncology teams can still adjust dosing or add otoprotective measures. Because the extended high-frequency band (8–16kHz) is affected first and standard audiograms rarely test above 8kHz, DPOAE is frequently the only clinical tool sensitive enough to catch ototoxic injury at its earliest, potentially most reversible stage.