Clinical assessment system and method of sound conduction in newborn ears using absorbance peak template

Abstract

An exemplary system and method are disclosed for the assessment of auditory sound conduction in the outer and middle ears in newborns and infants during the time of standard universal newborn hearing screening (UNHS) (inpatient and outpatient) in the first month of life using absorbance peak template (APT) analysis. Rather than using normative ranges showing the bounds of normal absorbance values at each frequency across frequencies, the exemplary system and method employ analysis of coherent morphological absorbance-frequency features, which are most diagnostically useful and theoretically informative, in the mid-frequency absorbance peak. The analysis can detect mid-frequency absorbance peaks in the absorbance peak template while also providing an overlay of the absorbance vs. frequency plot on the screen of the test instrument to determine key metrics in the parameters for normal absorbance peaks.

Description

Clinical Assessment System and Method of Sound Conduction in Newborn Ears using Absorbance Peak TemplateRELATED APPLICATION

[0001] This PCT application claims priority to, and the benefit of, U.S. Provisional Patent Application No. 63 / 637,115, filed April 22, 2024, entitled “Clinical Assessment System and Method of Sound Conduction in Newborn Ears using Absorbance Peak Template,” which is incorporated by reference herein in its entirety.BACKGROUND

[0002] Because temporary obstruction of the sound conduction pathway is the most common reason for failure on routine newborn hearing screening (NHS) tests, the incorporation of assessment of sound conduction at the time of NHS screening can improve screening outcomes.

[0003] Wideband immittance is a relatively new technology, emerging since the 1980s, and is not currently widely adopted in clinics despite commercial availability. However, the alternative routine test of immittance, tympanometry, is not appropriate for use in newborns due to immaturities in the ear canal wall that violate assumptions of tympanometry and result in inconsistent results. The feasibility of wideband immittance testing in newborns has been demonstrated as early as the 2000s. However, work was needed to further refine the measurements for clinical adoption.

[0004] There is a benefit to improving the sound conduction assessment of newborn ears.SUMMARY

[0005] An exemplary system and method are disclosed for the assessment of auditory sound conduction in the outer and middle ears in newborns and infants during the time of standard universal newborn hearing screening (UNHS) (inpatient and outpatient) in the first month of life using absorbance peak template (APT) analysis.

[0006] Absorbance peak template analysis can overcome challenges associated with middle ear wideband absorbance test technology, which can lead to improved outcomes for state-mandated Early Hearing Detection and Intervention (EHDI) programs; specifically, the ability to mitigate false-positive referrals that introduce inefficiencies (in resources and outcomes) to UNHS programs, and ultimately lead to delayed intervention and loss-to-follow- up of infants with permanent congenital hearing loss.

[0007] Rather than using normative ranges showing the bounds of normal absorbance values at each frequency across frequencies, the exemplary system and method employ analysis of coherent morphological absorbance-frequency features, which are most diagnostically useful and theoretically informative, in the mid-frequency absorbance peak. The analysis can detect mid-frequency absorbance peaks in the absorbance peak template while also providing an overlay of the absorbance vs. frequency plot on the screen of the test instrument to determine key metrics in the parameters for normal absorbance peaks.

[0008] Clinical study was conducted that developed and validated an absorbance peak template (APT) for the assessment of absorbance peaks in newborns. The study compared the test performance of absorbance peaks and APTs to existing normative methods to demonstrate APT-based methods for the categorization of abnormal absorbance peaks and to describe absorbance peak test-retest variability. The study concluded that the analysis of absorbance peaks guided by APT has the potential to simplify and improve assessments of sound conduction pathways in newborn ears and can be used together with or in place of current methods for analysis of wideband absorbance data.

[0009] The exemplary system and method can provide an analysis of a coherent morphological absorbance-frequency feature that can be readily identified and quantified, including visually by the test operator or automated by the machine. The exemplary system and method provide the ability to make inferences about the type of dysfunction made more objective, given specific parameters on normal peak frequency limits (e.g., shifting of absorbance peak to a higher frequency than normal, defined by APT, suggests increased stiffness; to a lower frequency suggests increased mass). Evidence that APT can detect smaller variations in measurements not discernable by existing methods allows for improved accuracy. Direct comparisons between these methods were reported in peer-reviewed publications.

[0010] In an aspect, a system (e.g., for analysis system) for screening abnormal sound conduction in a newborn or infant subject, the system comprising a processor; and a memory having instructions stored thereon, wherein when executed, the instructions cause the processor to receive wideband acoustic immittance (WAI) data including a wideband absorbance (WBA) measurement derived from acoustic reflectance responses recorded from an auditory canal of a subject in response to an application of a plurality of interrogative signals to the auditory channel by a piece of equipment comprising a vibration generation source (e.g., speaker, piezoelectric transducer, light vibrometer), wherein the plurality of interrogative signals includes signals at different primary frequency components in a broadband signal that includes a wide range of frequency components; determine a value for mid-frequency absorbance peakdefined as a maximum absorbance in the 750 Hz to 4000 Hz and a corresponding frequency of the maximum absorbance for a portion of the plurality of interrogative signals; and generate, via a display, (i) an absorbance frequency plot over a wide range of frequencies, including the determined value for the mid-frequency absorbance peak from the measurement and (ii) a graphical element comprising an absorbance peak template for the mid-frequency absorbance peak, wherein presence of a determined value for the mid-frequency absorbance peak in a region outside of the graphical element corresponding to the absorbance peak template and / or absence of a determined value for the mid-frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of the determined value for the mid-frequency absorbance peak in a region being inside the graphical element corresponding to the absorbance peak template indicates a normal sound conduction condition.

[0011] In some embodiments, the instructions (e.g., for the Absorbance maximum detection method) further cause the processor to apply a windowing function to one or more portions of the WAI data; and determine a value for a mid- frequency absorbance peak defined as a maximum absorbance in each windowed portion of the WAI data and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

[0012] In some embodiments, the instructions (e.g., Absorbance maximum detection method) further cause the processor to determine a second-derivative function of the WAI data; and determine, using the second-derivative function, a value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

[0013] In some embodiments, the WAI data further comprises an impedance magnitude and an impedance phase, wherein the instructions (Multi- variate, multi-stage method) further cause the processor to determine a value for an impedance minimum in the 750 Hz to 4000 Hz and a corresponding frequency of the impedance minimum for each of a portion of the plurality of interrogative signals; determine a frequency where the impedance phase is zero for a portion of the plurality of interrogative signals; and determine a value for mid-frequency absorbance peak defined as a maximum absorbance and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

[0014] In some embodiments, when the presence of a determined value for the midfrequency absorbance peak is in a region outside of the graphical element corresponding to the absorbance peak template or when a determined value for the mid- frequency absorbance peakis absent, the instructions (Automated classification of abnormal absorbance peak) further cause the processor to determine a classification of an abnormal sound conduction condition for the subject based on the location or absence of the mid-frequency absorbance peak.

[0015] In some embodiments, when the presence of a determined value for the midfrequency absorbance peak is in a region outside of the graphical element corresponding to the absorbance peak template or when a determined value for the mid- frequency absorbance peak is absent, the instructions (e.g., Automated classification of abnormal absorbance peak (with trained machine learning algorithm)) further cause the processor to determine, using a trained machine learning algorithm, a classification of an abnormal sound conduction condition for the subject based on the location or absence of the mid-frequency absorbance peak.

[0016] In some embodiments, the instructions further cause the processor to determine, using a trained machine learning algorithm, a value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

[0017] In some embodiments, the system further comprises a vibration generation source and a vibration recorder (e.g., microphone).

[0018] In some embodiments, the graphical element comprises two or more absorbance peak templates, wherein the presence of a determined value for the mid-frequency absorbance peak in a region outside of the graphical element corresponding to at least one of the absorbance peak templates or the absence of a determined value for the mid-frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein the presence of all of the determined value for the mid- frequency absorbance peak in a region being inside the graphical element corresponding to all absorbance peak templates indicates a normal sound conduction condition.

[0019] In some embodiments, the system is configured as an otoacoustic emissions (OAE) device.

[0020] In another aspect, a method is disclosed for screening abnormal sound conduction in a newborn or infant subject, the method comprising a) receiving wideband acoustic immittance (WAI) data comprising a wideband absorbance (WBA) measurement of measured reflectance responses acquired from an auditory canal of a subject in response to an application of a plurality of interrogative signals to the auditory channel by a piece of equipment comprising a vibration generation source (e.g., speaker, piezoelectric transducer, light vibrometer), wherein the plurality of interrogative signals includes signals at different primary frequency components; b) determining a value for mid-frequency absorbance peak defined asa maximum absorbance in the 750 Hz to 4000 Hz and a corresponding frequency of the maximum absorbance for each for a portion of the plurality of interrogative signals; and c) generating, via a display, (i) an absorbance frequency plot of the determined values for the midfrequency absorbance peak from the measurement and (ii) a graphical element comprising an absorbance peak template for the mid-frequency absorbance peak, wherein presence of a determined value for the mid-frequency absorbance peak in a region outside of the graphical element corresponding to the absorbance peak template or absence of a determined value for the mid-frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of all of the determined value for the mid- frequency absorbance peak in a region being inside the graphical element corresponding to the absorbance peak template indicates a normal sound conduction condition.

[0021] In some embodiments, the step b) further comprises applying a windowing function to one or more portions of the WAI data; and determining a value for a mid-frequency absorbance peak defined as a maximum absorbance in each windowed portion of the WAI data and a corresponding frequency of the maximum absorbance for each for a portion of the plurality of interrogative signals.

[0022] In some embodiments, the step b) further comprises determining a second- derivative function of the WAI data; and determining, using the second-derivative function, a value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

[0023] In some embodiments, the WAI data further comprises an impedance magnitude and an impedance phase, and the step b) further comprises determining a value for an impedance minimum in the 750 Hz to 4000 Hz and a corresponding frequency of the impedance minimum for each of a portion of the plurality of interrogative signals; determining a frequency where the impedance phase is zero for a portion of the plurality of interrogative signals; and determining a value for mid- frequency absorbance peak defined as a maximum absorbance and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

[0024] In some embodiments, the step b) further comprises determining, when the presence of a determined value for the mid-frequency absorbance peak is in a region outside of the graphical element corresponding to the absorbance peak template or when a determined value for the mid-frequency absorbance peak is absent, a classification of an abnormal soundconduction condition for the subject based on the location or absence of the mid-frequency absorbance peak.

[0025] In some embodiments, the step b) further comprises determining, when the presence of a determined value for the mid-frequency absorbance peak is in a region outside of the graphical element corresponding to the absorbance peak template or when a determined value for the mid-frequency absorbance peak is absent, using a trained machine learning algorithm, a classification of an abnormal sound conduction condition for the subject based on the location or absence of the mid-frequency absorbance peak.

[0026] In some embodiments, the step b) is executed using a trained machine learning algorithm.

[0027] In some embodiments, the graphical element comprises two or more absorbance peak templates, wherein the presence of a determined value for the mid-frequency absorbance peak in a region outside of the graphical element corresponding to at least one of the absorbance peak templates or the absence of a determined value for the mid-frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein the presence of all of the determined value for the mid- frequency absorbance peak in a region being inside the graphical element corresponding to all absorbance peak templates indicates a normal sound conduction condition.

[0028] In another aspect, a method for using wideband absorbance in a hearing screening (NHS) procedure for a newborn or infant subject is disclosed comprising receiving a first wideband acoustic immittance (WAI) data including a wideband absorbance (WBA) measurement derived from reflectance responses recorded from an auditory canal of the subject in response to an application of a plurality of interrogative signals to an auditory channel by a piece of equipment including a vibration generation source, wherein the plurality of interrogative signals includes signals at different primary frequency components in a broadband signal that includes a wide range of frequency components; and in response to the first WAI data failing a first test performed by an otoacoustic emissions (OAE) device or an automated auditory brainstem response (AABR) device: determining a first value for midfrequency absorbance peak defined as a maximum absorbance in a 750 Hz to 4000 Hz range and a first corresponding frequency of the maximum absorbance for a portion of the plurality of interrogative signals, and generating, via a display, (i) a first absorbance frequency plot over a wide range of frequencies, including the first determined value for the mid-frequency absorbance peak from the measurement and (ii) a graphical element including an absorbance peak template for the mid-frequency absorbance peak, wherein presence of a first determinedvalue for the mid-frequency absorbance peaks in a region outside of the graphical element corresponding to the absorbance peak template or absence of a determined value for the midfrequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of the first determined value for the mid-frequency absorbance peak in a region being inside the graphical element corresponding to the absorbance peak template indicates a normal sound conduction condition.

[0029] In some embodiments, the method described herein further comprises: in response to the subject being determined as having the normal sound conduction condition, outputting the first absorbance frequency plot and / or the first determined value for the mid-frequency absorbance peak as a first data object, wherein the outputted first data object is subsequently employed in monitoring or diagnostics devices (e.g., for prioritized result-based auditory brainstem response (ABR DX) monitoring).

[0030] In some embodiments, the method described herein further comprises: in response to the subject being determined as having the abnormal sound conduction condition: receiving a second wideband acoustic immittance (WAI) data derived from reflectance responses recorded from the auditory canal of the subject in response to the application of the plurality of interrogative signals to the auditory channel by the piece of equipment including the vibration generation source, wherein the plurality of interrogative signals includes signals at different primary frequency components in the broadband signal that includes the wide range of frequency components, and in response to the second WAI data failing a second test performed by an otoacoustic emissions (OAE) device or an automated auditory brainstem response (AABR) device: determining a second value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz range and a second corresponding frequency of the maximum absorbance for a portion of the plurality of interrogative signals, and generating, via a display, a second absorbance frequency plot over a wide range of frequencies, including the second determined value for the mid-frequency absorbance peak from the measurement, wherein presence of a second determined value for the mid-frequency absorbance peaks in a region outside of the graphical element corresponding to the absorbance peak template or absence of a determined value for the mid-frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of the second determined value for the mid-frequency absorbance peak in a region being inside the graphical element corresponding to the absorbance peak template indicates a normal sound conduction condition.

[0031] In some embodiments, the method described herein further comprises: in response to the first WAI data passing the test performed by the otoacoustic emissions (OAE) device or the automated auditory brainstem response (AABR) device: determining one or more risk factors (e.g., infections) relating to the hearing of the subject; and outputting one or more risk factors as a second data object, wherein the outputted second data object is subsequently employed in screening or diagnostic devices (e.g., for ABR Dx monitoring).

[0032] In some embodiments, the method described herein further comprises: in response to the subject being determined as having the normal sound conduction condition after the second absorbance frequency plot is generated, outputting the second absorbance frequency plot and / or the second determined value for the mid-frequency absorbance peak as a third data object, wherein the outputted third data object is subsequently employed in screening or diagnostics devices (e.g., for prioritized ABR Dx monitoring).

[0033] In some embodiments, the method described herein further comprises: in response to the subject being determined as having the abnormal sound conduction condition after the second absorbance frequency plot is generated, outputting the second absorbance frequency plot and / or the second determined value for the mid-frequency absorbance peak as a fourth data object, wherein the outputted fourth data object is subsequently employed for an outpatient rescreening procedure.

[0034] Other systems, methods, features, and / or advantages will be or may become apparent to one with skill in the art upon examination of the following drawings and detailed description. It is intended that all such additional systems, methods, features and / or advantages be included within this description and be protected by the accompanying claims.BRIEF DESCRIPTION OF DRAWINGS

[0035] Figs. 1A and IB each show an example system configured to assess auditory sound conduction in the outer and middle ears in newborns and infants during the time of standard universal newborn hearing screening (UNHS) (inpatient and outpatient) using absorbance peak template (APT) analysis in accordance with an illustrative embodiment.

[0036] Figs. 2A - 2C show example operation flows for the exemplary system to determine whether the determined value for the mid-frequency absorbance peaks is within the absorbance peak template and / or outside of the absorbance peak template. Figs. 2A - 2B each shows an operation flow using second-derivative functions to determine the peak of absorbance as a function of frequency, combinable with a windowing function to assay maximum values atdifferent frequency intervals. Fig. 2C shows another operation flow using Multivariate multistage analysis.

[0037] Figs. 3A - 3F show example characteristics in newborn hearing screenings and newborn hearing screening (NHS) protocols. Fig. 3A shows example characteristics of the WBA normative range and absorbance area index in newborns. Fig. 3B is a visualization showing a low-frequency peak, a middle-frequency peak, and a high-frequency peak used in the absorbance assessment. Fig. 3C shows example wideband absorbance immittance (WAI) measures (from which WBA is derived) as power absorbance, impedance magnitude, and impedance phase. Fig. 3D shows an example newborn hearing screening (NHS) protocol employing the wideband absorbance (WBA) test with an APT. Fig. 3E shows an example newborn hearing screening (NHS) protocol without the wideband absorbance (WBA) test. Fig. 3F shows an example result for the WBA assessment of the sound conduction pathway using the APT.

[0038] Figs. 4A - 4B shows the evaluation results for a wideband absorbance (WBA) assessment using an absorbance peak template (APT). Fig. 4A shows a scatter plot of peak absorbance versus peak frequency from 490 analyzed peaks, wherein the APT is plotted as a rectangular shape. Fig. 4B shows a comparison between (i) APT-based assessment of absorbance peaks and (ii) assessment of wideband absorbance across frequency using the traditional normative range.DETAILED DESCRIPTION

[0039] It is appreciated that certain features of the disclosure, which are, for clarity, described in the context of separate aspects, can also be provided in combination with a single aspect. Conversely, various features of the disclosure, which are, for brevity, described in the context of a single aspect, can also be provided separately or in any suitable subcombination. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. Methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present disclosure.

[0040] Example System

[0041] Figs. 1A and IB show an example system 100 (shown as 100a, 100b) configured to assess auditory sound conduction in the outer and middle ears (shown as auditory canal 101) in newborns and infants during the time of standard universal newborn hearing screening (UNHS) (inpatient and outpatient) using absorbance peak template (APT) analysis (i.e.,absorbance frequency plot 103) in accordance with an illustrative embodiment. In Fig. 1 A, the system 100a is configured with an analysis 110 module configured to perform the absorbance peak template (APT) analysis on the measurement instrument. In Fig. 1 A, the system 100b is configured with an analysis 110 module external to the measurement instrument, e.g., on a local computing device in operative (e.g., network connection) with the measurement instrument or a remote computing device in the cloud (e.g., cloud infrastructure) in operative network connection with the measurement instrument or an edge device connected to the measurement instrument. The analysis 1 10 module may be a part of an auditory assessment software package, e.g., provided by the manufacturer of the measurement instrument or a software package separately provided by the instrument manufacturer.

[0042] In the example shown in Fig. 1A and IB, the system 100 (e.g., 100a, 100b) is configured to apply interrogative signals 105 (shown as forward sound) to the auditory channel 101 (shown as the auditory canal of the ear) using a speaker 102. The interrogative signals 105, or a portion thereof, can be absorbed by the eardrum in the middle ear (shown as absorbed acoustic energy 109) and partially reflected back to a microphone 104 as reflectance responses 107 (shown as reflected sound) to provide a wideband absorbance (WBA) measurement of the reflectance responses 107 (shown as reflected sound) as a measure of absorbed acoustic energy. In some embodiments, the speaker 102 and the microphone 104 can be integrated into a microphone device 111 configured to generate forward sound 105 and receive reflected sound 107. Front-end electronics 106 is configured to receive the received reflectance sound 107 (shown as reflected sound) to generate a wideband absorbance (WBA) measurement signal as a measure of absorbed acoustic energy. The signals can be applied, e.g., as tones ranging from 20 Hz to 8000 Hz. In some embodiments, the wideband absorbance measurement is applied as a chirp having multiple frequency components.

[0043] In Fig. 1 A, the controller 108 is configured to receive wideband acoustic immittance (WAI) data comprising a wideband absorbance (WBA) measurement. The controller 108 includes an analysis component (shown as analyzer 110) configured to determine a value for mid- frequency absorbance peak 113 having a maximum absorbance in the 750 Hz to 4000 Hz (shown as range 115) and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals 105 to which used by the absorbance peak template (APT) analysis. In some embodiments, e.g., in Fig. IB, the controller 108 or instrument device (shown as 110’) is configured to transmit, via a device interface 114 or other interface, the determined mid- frequency absorbance peak 113 (and its associated maximum absorbance and corresponding frequency of the maximum absorbance) as an input data to adiagnosis system or application in the cloud or local computing device. The controller 108 or device can, in some embodiments, provide the measurement and absorbance peak template to a patient portal, etc.).

[0044] The absorbance peak template may be shown via a visualization or rendering of the analysis software from the analysis module 110, which defines the peak absorbance reading at 95% and 5% and the peak frequency reading at 5% and 95%. Fig. IB shows an alternative visualization of the absorbance measurement with the absorbance peak template. The system of Fig. 1 A may output a similar alternative visualization of the absorbance measurement with the absorbance peak template. In Fig. IB, the software output includes the absorbance peak template 117 visualization and additional visualization 119 correspondence to normal curves for a certain population set. The visualization 119 (shaded area) may be stored for certain population groups based on age (e.g., newborns, older infants, or children) and presented based on a selectable drop-down menu.

[0045] Notably, the absorbance peak template 117 provides a tangible visual output having a clinical utility that can assist a clinician / doctor in making an assessment that the patient (e.g., newborn) had a normal conductive hearing response or has temporary conductive hearing loss to which additional follow-up may be prescribed.

[0046] Currently, early hearing detection systems have a large referral volume of predominantly temporary hearing loss, i.e., conductive hearing loss that is only temporary and that is currently flooding or being a bottleneck in the system. These are OAEs or ABRs, physiological tests that either test the inner ear or the neural system and when they fail currently, clinicians don't know why they fail - they could be temporary or conductive hearing loss. A large portion of the cases are temporary (e.g., greater than 70%), e.g., due to vemix and amniotic fluid in the ear, and such cases, to which a quick follow-up response would have the benefit of removing such cases from the healthcare system. The absorbance peak template 1 17 and the associated diagnostic system can facilitate the early screening of such issues and their quick removal - a substantial improvement to the healthcare system (in terms of cost and expanding availability of limited resources to cases that warrant them) and to the patient (in term of having earlier treatment due to earlier identification of conductive hearing loss). In the 30% of remaining cases, absorbance peak template 117 and associated systems can rule out the presence of temporary conductive hearing loss and provide them with direct access to priority diagnostic care rather than conducting several follow-up visits on a wait-and-rescreen basis.

[0047] In Fig. 1A, the controller 108 can generate, via a display 112’ (using display interface 112), (i) an absorbance frequency plot 103 of the determined values for the mid-frequency absorbance peaks from the measurement, and (ii) a graphical element comprising an absorbance peak template 117 for the mid-frequency absorbance peaks. The presence of a determined value for the mid-frequency absorbance peaks 113 in a region outside of the graphical element 117 corresponding to the absorbance peak template and / or the absence of a determined value for the mid- frequency absorbance peaks 113 indicates an abnormal sound conduction condition for the subject. The presence of all of the determined values for the midfrequency absorbance peaks 113 in a region inside the graphical element 117 corresponding to the absorbance peak template indicates a normal sound conduction condition.

[0048] Wideband absorbance (WB A) is one measure derived from ear canal sound pressure recordings. Using signal processing and mathematical computations, one can derive additional measures belonging to the broader class of this test technology, called wideband acoustic immittance (WAI). For instance, the impedance measured across frequencies, impedance phase, or analogous admittance or admittance phase across frequency is often saved using the same recordings that produce WBA. These measures are mathematically related, but unlike absorbance, the other measures are not considered for clinical application.

[0049] The exemplary system and method may be performed in the context of a clinical protocol, e.g., described in relation to Figs. 3E and 3F.

[0050] Example Method

[0051] Figs. 2A - 2C show example operation flows 200a - 200c for the exemplary system to determine whether the determined value for the mid-frequency absorbance peaks (shown as 1 13 in Figs. 1A and IB) is within the absorbance peak template and / or outside of the absorbance peak template (shown as 117 in Fig. 1).

[0052] Absorbance maximum detection method #1. This method could use second- derivative functions to determine the peak of Absorbance as a function of frequency, denoted as A(f). This could be combined with a windowing function that assays maximum values at different frequency intervals between 750 Hz to 4000 Hz (range 115 in Fig. 1) and beyond.

[0053] Fig. 2A shows the operation flow 200a for the determination of the peak of Absorbance. At step 202, the exemplary system is configured to apply a windowing function to one or more portions of the WAI data. At step 204, the exemplary system is configured to determine a value for a mid-frequency absorbance peak defined as a maximum absorbance in each windowed portion of the WAI data and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

[0054] Fig. 2B shows additional, or alternative, operation flow 200b that may be performed for the determination of the peak of Absorbance. At step 206, the exemplary system candetermine a second-derivative function of the WAI data. At step 208, the exemplary system can determine, using the second-derivative function, a value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz range and a corresponding frequency of the maximum absorbance for each portion of the plurality of interrogative signals. In some embodiments, both assessments may be performed to which an average may be derived and presented as the output.

[0055] Multi-variate, multi-stage method. If more precision is desired, assays of impedance magnitude and impedance phase can be employed in which A) Impedance minima are identified on the impedance vs. frequency function Z (f) using a similar second-derivative formula as #1, together with identifying frequency values of impedance phase = 0. (B) Once the frequency value is determined, a honed-in assay of WBA will focus on a narrower frequency interval to detect and measure the value [size] of the absorbance peak within that interval.

[0056] Fig. 2C shows another operation flow 200c for the determination of the peak of Absorbance based on Multi-variate, multi-stage analysis, which can comprise 3 steps. At step 210, the exemplary system can determine a value for an impedance minimum in the 750 Hz to 4000 Hz range and a corresponding frequency of the impedance minimum for each portion of the plurality of interrogative signals. At step 212, the exemplary system can determine a frequency where the impedance phase is zero for a portion of the plurality of interrogative signals. At step 214, the exemplary system can determine a value for the mid-frequency absorbance peak defined as a maximum absorbance and a corresponding frequency of the maximum absorbance for each portion of the plurality of interrogative signals.

[0057] Example Machine Learning-Based Detection Model

[0058] Modeling and machine learning paradigms. The behavior of absorbance peaks, impedance minima, and phase values of ‘0” may lead to model-driven methods for accurate detection. Though the mathematical relationship is understood among those variables, (A) signal processing models can account for / minimize the error or impact of noise in data that could introduce inaccuracies in the detection of peaks. (B) machine-learning paradigms are being employed in adults for feature detection, e.g., the width of the notch between two peaks for cases of superior canal dehiscence. Employing a machine learning paradigm may be a simpler task for absorbance peaks in newborns, given peak is simpler in its morphology and our vast training and validation data.

[0059] Indeed, in some embodiments, a machine learning classifier is employed to detect an abnormal sound conduction condition for the subject based on the location or absence of themid- frequency absorbance peaks. Examples of machine learning operations that may be used are provided herein.

[0060] Annotation labels for the machine learning paradigms. To train the ML-based detection model of the exemplary system to determine if an absorbance peak is within the APT or not, a training system can use a training data set annotated with (i) the status of acoustic leaks and ear canal tip fit, (ii) the status of collapsing ear canal walls (e.g., suspected, collapsed, open), and / or (iii) noise or incomplete sequence of WBA testing as the training annotation (i.e., label) for the detection model.

[0061] A study was conducted that acquired absorbance peaks from measurements from 297 newborns and a total of 484 ear canals. In each measurement, absorbance peaks were picked manually by trained research assistants. These measurements were also validated for the presence of acoustic leaks via the ability or inability to obtain a hermetic seal and an assay of collapsed ear canal walls using pressure-induced procedures to collapse the flaccid canal walls and collapse them, leading to the identification of WBA patterns that characteristically indicate collapsed canals. Subsequently, simple methods to identify mid-frequency absorbance peaks, e.g., using the excel maxima detection algorithm, failed to account for artificial maxima, artificial increase in absorbance, or rising / falling slopes of absorbance across frequency in the presence of acoustic leaks and / or collapsed canals. As a result, a training data set for the ML- based detection model included in addition to manually identified absorbance peaks, coding variables that specify the presence of / or suspicion of acoustic leaks or spontaneously collapsed ear canal walls.

[0062] Machine Learning. The term “artificial intelligence” can include any technique that enables one or more computing devices or computing systems (i.e., a machine) to mimic human intelligence. Artificial intelligence (Al) includes but is not limited to knowledge bases, machine learning, representation learning, and deep learning. The term “machine learning” is defined herein to be a subset of Al that enables a machine to acquire knowledge by extracting patterns from raw data. Machine learning techniques include, but are not limited to, logistic regression, support vector machines (SVMs), decision trees, Naive Bayes classifiers, and artificial neural networks. The term “representation learning” is defined herein to be a subset of machine learning that enables a machine to automatically discover representations needed for feature detection, prediction, or classification from raw data. Representation learning techniques include, but are not limited to, autoencoders. The term “deep learning” is defined herein to be a subset of machine learning that enables a machine to automatically discover representations needed for feature detection, prediction, classification, etc., using layers of processing. Deeplearning techniques include but are not limited to artificial neural networks or multilayer perceptron (MLP).

[0063] Machine learning models include supervised, semi-supervised, and unsupervised learning models. In a supervised learning model, the model learns a function that maps an input (also known as feature or features) to an output (also known as target or target) during training with a labeled data set (or dataset). In an unsupervised learning model, the model has a pattern in the data. In a semi-supervised model, the model learns a function that maps an input (also known as a feature or features) to an output (also known as a target) during training with both labeled and unlabeled data.

[0064] Neural Networks. An artificial neural network (ANN) is a computing system including a plurality of interconnected neurons (e.g., also referred to as “nodes”). This disclosure contemplates that the nodes can be implemented using a computing device (e.g., a processing unit and memory as described herein). The nodes can be arranged in a plurality of layers, such as an input layer, an output layer, and optionally one or more hidden layers. An ANN having hidden layers can be referred to as a deep neural network or multilayer perceptron (MLP). Each node is connected to one or more other nodes in the ANN. For example, each layer is made of a plurality of nodes, where each node is connected to all nodes in the previous layer. The nodes in a given layer are not interconnected with one another, i.e., the nodes in a given layer function independently of one another. As used herein, nodes in the input layer receive data from outside of the ANN, nodes in the hidden layer(s) modify the data between the input and output layers, and nodes in the output layer provide the results. Each node is configured to receive an input, implement an activation function (e.g., binary step, linear, sigmoid, tanH, or rectified linear unit (ReLU) function), and provide an output in accordance with the activation function. Additionally, each node is associated with a respective weight. ANNs are trained with a dataset to maximize or minimize an objective function. In some implementations, the objective function is a cost function, which is a measure of the ANN’s performance (e.g., an error such as LI or L2 loss) during training, and the training algorithm tunes the node weights and / or bias to minimize the cost function. This disclosure contemplates that any algorithm that finds the maximum or minimum of the objective function can be used for training the ANN. Training algorithms for ANNs include but are not limited to backpropagation. It should be understood that an artificial neural network is provided only as an example machine learning model. This disclosure contemplates that the machine learning model can be any supervised learning model, semi-supervised learning model, or unsupervisedlearning model. Optionally, the machine learning model is a deep learning model. Machine learning models are known in the art and are therefore not described in further detail herein.

[0065] A convolutional neural network (CNN) is a type of deep neural network that has been applied, for example, to image analysis applications. Unlike traditional neural networks, each layer in a CNN has a plurality of nodes arranged in three dimensions (width, height, and depth). CNNs can include different types of layers, e.g., convolutional, pooling, and fully- connected (also referred to herein as “dense”) layers. A convolutional layer includes a set of filters and performs the bulk of the computations. A pooling layer is optionally inserted between convolutional layers to reduce the computational power and / or control overfitting (e.g., by downsampling). A fully connected layer includes neurons, where each neuron is connected to all of the neurons in the previous layer. The layers are stacked similarly to traditional neural networks. GCNNs are CNNs that have been adapted to work on structured datasets such as graphs.

[0066] Other Supervised Learning Models. A logistic regression (LR) classifier is a supervised classification model that uses the logistic function to predict the probability of a target, which can be used for classification. LR classifiers are trained with a data set (also referred to herein as a “dataset”) to maximize or minimize an objective function, for example, a measure of the LR classifier’s performance (e.g., an error such as LI or L2 loss), during training. This disclosure contemplates that any algorithm that finds the minimum of the cost function can be used. LR classifiers are known in the art and are therefore not described in further detail herein.

[0067] A Naive Bayes’ (NB) classifier is a supervised classification model that is based on Bayes’ Theorem, which assumes independence among features (i.e., the presence of one feature in a class is unrelated to the presence of any other features). NB classifiers are trained with a data set by computing the conditional probability distribution of each feature given a label and applying Bayes’ Theorem to compute the conditional probability distribution of a label given an observation. NB classifiers are known in the art and are therefore not described in further detail herein.

[0068] A k-NN classifier is a supervised classification model that classifies new data points based on similarity measures (e.g., distance functions). The k-NN classifiers are trained with a data set (also referred to herein as a “dataset”) to maximize or minimize a measure of the k- NN classifier’s performance during training. The k-NN classifiers are known in the art and are therefore not described in further detail herein.

[0069] A majority voting ensemble is a meta-classifier that combines a plurality of machine learning classifiers for classification via majority voting. In other words, the majority voting ensemble’s final prediction (e.g., class label) is the one predicted most frequently by the member classification models. The majority voting ensembles are known in the art and are therefore not described in further detail herein.

[0070] Automated classification of abnormal absorbance peaks: Another aspect of automated detection and analysis is the ability to assess the presence or absence of the peak and, when outside the APT graphical element, to classify it into the assessment of categories, e.g., frequency of peak to the right of the APT, or the left of APT, or below the APT. In the paper, these categories correspond to A (within APT), B (to the right or left of APT), and C&D, which will need a new unified category name (e.g., just C).

[0071] Analysis of Absorbance Peaks

[0072] Absorbance peak analysis, as used herein, targets the mid-frequency absorbance peak, which can be defined as the peak in the 750-to 4000-Hz frequency range. The analysis range can be selected to detect normal variation in the mid-frequency peak while avoiding the lower frequency absorbance peak (typically 300-500 Hz) and the higher frequency peak (around 6000 Hz). The analysis may include determining (a) the maximum value of absorbance in the 750-Hz to 4000-Hz range, i.e., “peak absorbance,” and (b) the frequency point at which the maximum occurred, i.e., “peak frequency.”

[0073] Multivariate analysis. The discriminant function analysis may combine both peak variables, peak absorbance, and peak frequency, into a single multivariate quantity called the discriminant function score (DFS). The peak variables (independent variables) predicted the DPOAE pass / fail group membership (dependent variable). The discriminant function coefficients for each of the independent variables may convey their relative importance in the prediction of DPOAE pass / fail outcomes. Also, the resulting coefficients and intercept (constant) values were used to construct a DFS equation that can be used to reproduce the DFS scores based on raw values of the independent variables.

[0074] Absorbance area indices. Fig. 3A shows example characteristics of the WBA normative range and absorbance area index (AAI) in newborns. Absorbance area indices (AAIs) refer to averages of absorbance over specified frequency intervals. The computation of AAI in this manner is an adaptation of the RAIs concept first described by Hunter et al. (2010) and used by others (S. Aithal et al., 2015). AAIs may be computed using absorbance from the validation set over seven frequency intervals: 200-6000, 1000-2000, 1000-4000, 1000-6000, 2000-4000, and 4000-6000 Hz. These frequency intervals were selected so that our AAIs arecomparable to RAIs in the literature. Using ROC analyses, the test performance of these AAIs, indicated by their respective AROC values, was compared to those of absorbance peak variables and the multivariate DFS variable.

[0075] Absorbance peaks. Absorbance peaks at specific frequencies can differentiate between normal and impaired hearing in newborns and infants. For example, variations in absorbance at frequencies 2, 3, and 6 KHz have been observed when comparing different newborn hearing results in previous studies. Absorbance peaks analysis categorizes three types of absorbance peaks, including (i) low-frequency (low- / ) peak (i.e., ear canal wall resonance), (ii) middle-frequency (mid- / ) peak (i.e., middle ear resonance), and (iii) high-frequency (high- / ) peak (i.e., tympanic cavity resonance). Fig. 3B is a visualization showing a low- - peak 302, a mid- / ' peak 304, and a high- / ' peak 306.

[0076] To get a clear reflectance response (i.e., shown as 107 in Fig. 1) from the middle ear to generate WBA measurement, the applied forward sound (shown as 105 in Fig. 1) should have a frequency (denoted as / ) at which an impedance (denoted as Z) (i.e., physical property that hinders sound vibration of the eardrum) of the middle ear is the smallest. In other words, the middle ear reflects sound best (i.e., best reflectance response) at its natural (resonant) frequency / ’which results in the smallest impedance Z, which can be defined per Equation 1.(Eq. 1)

[0077] In Equation 1 , f is the frequency of the forward sound (shown as 105 in Fig. 1), k is the stiffness of the eardrum and ligaments supporting the ossicles, M is the mass of the ossicles and surrounding bones, and R is the resistance caused by friction and damping within the middle ear.

[0078] Wideband acoustic immittance (WAI). The wideband absorbance measurement, indicating mid-frequency absorbance peak and corresponding frequency of the peak, can be retrieved from wideband acoustic immittance (WAI) data, which can be a measure for power absorbance, impedance magnitude, phase, or power reflectance. Equation 2 defines the WAI measure as a power absorbance. Fig. 3C shows example WAI measures as power absorbance, impedance magnitude (i.e., Z magnitude), and impedance phase (i.e., Z phase).Absorbed powerPower absorbance = - - - -Incident power(Eq. 2)

[0079] Example Newborn Hearing Screening (NHS) Protocol

[0080] Fig. 3D shows an example newborn hearing screening (NHS) protocol employing the wideband absorbance (WBA) test with an APT. In subpanel (a), at step 310, a newborn’s ears can be screened using an otoacoustic emission (OAE) and / or an automated auditory brainstem response (AABR) test to determine if they have hearing loss. If the newborn's ears pass the OAE or AABR test and reach step 312, further analysis of potential risk factors for the ears is carried out. When no risk factors are found, the NHS protocol can be discharged. When potential risk factors are found, the NHS protocol can employ further monitoring mechanisms of the ears (e.g., ABR Dx).

[0081] If the newborn’s ears fail the OAE or AABR test at step 310 (i.e., they experience hearing loss), the WBA test can be performed on the newborn’s ears at step 314 (shown as 314’), using the APT template, to generate a WBA measurement. If the WBA measurement has a mid-frequency peak centered within the APT (shown as normal absorbance at step 316), the ears can be concluded as experiencing permanent non-conductive hearing loss, and they can be prioritized for diagnostic evaluation at step 320, using result-based auditory brainstem response (ABR Dx) monitoring. If the WBA measurement has a mid-frequency peak outside the APT (shown as abnormal absorbance at step 318), the ears can be concluded as experiencing temporary conductive hearing loss and further examined in a second (repeated) OAE or AABR test 310’. The protocol can repeat the process via the same operation using the OAE / AABR test 310’ and WBA test 314’.

[0082] In some embodiments, shown in subpanel (b), the NHS protocol employing WBA assessment only carries out one WBA test 314 if the newborn's ears fail the OAE or AABR screening 310. The NHS protocol does not repeat the screening process as in subpanel (a).

[0083] Fig. 3E shows an example newborn hearing screening (NHS) protocol without the wideband absorbance (WBA) test. Subpanel (a) shows an early detection hearing intervention (EHDI) 1-3-6 phases of the protocol comprising (i) a 1-month identification phase, (ii) a 3- month ABR Dx phase, and (iii) a 6-month intervention phase. The entire EHDI process may last up to 10 months, providing a delayed hearing loss detection and intervention. Subpanel (b) shows the current state-of-the-art NHS protocol without the WBA test, including (i) one- technology NHS protocol using either OAE or AABR test, and (ii) two-tier NHS protocol using OAE and AABR tests, wherein the OAE and AABR screenings have a high failure rate (e.g., 76-92%) due to temporary dysfunction in the sound conduction pathway of the newborn at birth.

[0084] Combining the WBA assessment with the NHS protocol can mitigate the impact of high failure rates through an improved ref erral / decision- making matrix (AlMakadma et al., 2023b) shown in Table 3. The ability given by the additional assessment of WBA measures at the time of NHS to discern in some cases whether a 'fail' outcome is due to cochlear function provides conclusive identification of cases at high risk of permanent non-conductive hearing loss. For such cases, referral paradigms should prioritize diagnostic ABR evaluation and eliminate unnecessary follow-up re-screens.Table 3

[0085] Fig. 3F shows an example result for the WBA assessment of the sound conduction pathway using APT. As shown, normal WBA measurements have their mid-frequency peak centered within the absorbance peak template (APT), indicating normal absorbance value and normal frequency of the peak. Abnormal absorbance peaks are frequency-shifted outside of the APT or reduced below it.

[0086] Experimental Results and Additional Examples

[0087] A study was conducted that developed and validated an absorbance peak template (APT) for an assessment of absorbance peaks in newborns. The study compared the test performance of absorbance peaks and APTs to existing normative methods to demonstrate APT-based methods for categorization of abnormal absorbance peaks and to describe absorbance peak test-retest variability. The study concluded that the analysis of absorbance peaks guided by APT has the potential to simplify and improve assessments of sound conduction pathways in newborn ears and can be used together with or in place of current methods for analysis of wideband absorbance data.

[0088] Assessment of absorbance peaks resulted in test performance similar to that of current methods for analysis of wide-band absorbance measurements. The APT evaluated in the study provides a straightforward paradigm for the assessment of absorbance peaks. Given its defined boundaries for normal measurements and simple categories for assessment of abnormal absorbance peaks, APT-based analysis is also amenable to automation, which is ideal for use in newborn hearing screening settings. Moreover, the findings suggested that APT-based analyses may allow for the detection of subtle changes in the sound conduction pathway that are otherwise not possible with current normative paradigms.

[0089] A study was conducted that contemplated and annotated wide-band absorbance measurements for use in ML / AI training; the annotation included (i) the status of acoustic leaks and ear canal tip fit, (ii) the status of collapsing ear canal walls (e.g., suspected, collapsed, open), and / or (iii) noise or incomplete sequence of WBA testing as the training annotation (i.e., label) for the detection model.

[0090] Experiment Procedure

[0091] Preparation of training and validation datasets. The study used a training dataset, detailed in AlMakadma and Prieve (2021), to develop an absorbance peak template (APT) in newborns with normal sound conduction (in the outer / middle ears). Newborns younger than 48 hours underwent repeated trials of transient evoked OAE (TEOAE) and WAI testing, with probe removal and reinsertion between each trial. Testing was conducted in each ear using the HearlD system (Mimosa Acoustics Inc.) for one set of repeated trials (M= 3.2 ± 1.5 insertions per ear) and using the Titan system (Interacoustics A / S) for another set of trials (M = 5.6 ± 2.6 insertions per ear). The study counterbalanced the order of testing between left and right ears and between the HearlD and Titan systems. A previous study (e.g., AlMakadma and Prieve) analyzed this data to develop probe-fit criteria to control for artificial inflation in absorbance due to loose probe fits / acoustic leaks. The average artificial inflation was 0.1 at frequencies between 1000 and 6000 Hz, where the mid- frequency absorbance peak occurred in newborns. At the end of the analysis, the previous study determined that all measurements with low- / absorbance (averaged over 250-1000 Hz) > 0.58 and low- / impedance phase (averaged over 500-1000 Hz) > -0.11 cycles should be excluded from the data on suspension of loose / leaky fits. Following the exclusion, 490 absorbance measurements from 84 ears remained with no significant leak-related inflation, 72 of which had measurements repeated using both systems (AlMakadma & Prieve, 2021). These measurements constituted the training set for the instant study.

[0092] The study used a separate dataset, detailed in Sanford et al. (2009), for validation analyses to assess the performance of the peak template in classifying ears with normal sound conduction and ears with abnormal sound conduction, using the newborn screening outcomes as a reference standard. The data in the dataset were collected from 455 newborns who received a distortion-product OAE (DPOAE) hearing screening test. The study conducted wideband tympanometry, 1-kHz tympanometry, and WAI testing in the ambient condition in 375 ears that passed the DPOAE screening (average age = 25 ± 8 h) and in 80 ears that failed thescreening (average age = 22 + 8 h). The study performed the wideband tympanometry and WAI tests using the same probe insertion and a prototype Interacoustics system. For newborns who failed their initial DPOAE test (on Day 1), the study performed repeated screens and testing the following day (Day 2). Only absorbance measurements from (ambient) WAI testing performed on Day 1 were retained for the present investigation. A difference from the training set is that Sanford et al. utilized an automated system alert that indicated to the operator whenever a leak was present, prompting a refit and retest. Specifically, the system determined a leak was present whenever ]ow-f equivalent volume (averaged over 500-1000 Hz) < -1.15 cm3or if any 1 / 2-octave equivalent volume minima between 500 and 1000 Hz were < -2.3 cm3.

[0093] The study applied additional refinement of the validation data set to ensure that leaks and other sources of artifact / error did not affect absorbance peaks, which was consistent with the methods applied to the training set. AlMakadma and Prieve (2021) demonstrated that a low- / equivalent volume probe-fit criterion, such as applied by Sanford et al. (2009), was insufficient to rule out inflation in absorbance, including in the absorbance peak frequency range (between 1000 and 2000 Hz). Therefore, the study also applied the probe fit criteria recommended by AlMakadma and Prieve (2021) to the validation set in the same manner as the training set, resulting in the exclusion of 20 absorbance measurements. The study excluded an additional 12 absorbance measurements for having values < 0. Such errors in measurement may be associated with acoustic termination of the probe, for example, with occlusion against the ear canal wall, collapsed ear canals, and / or in relationship to calibration errors (AlMakadma, Aithal, et al., 2023). As a result, the remaining 359 measurements from ears that passed the DPOAE screening and 64 measurements from ears that failed the screening were retained for validation analysis in the study.

[0094] Analysis of absorbance peaks. Absorbance peak analysis in the study targeted the mid-frequency absorbance peak, which was defined as the peak in the 750- to 4000-Hz frequency range. This analysis range was selected to detect normal variation in the midfrequency peak while avoiding the lower frequency absorbance peak (e.g., 300-500 Hz) and the higher frequency peak (e.g., around 6000 Hz). In this analysis, the study determined two peak variables: (a) the maximum value of absorbance in the 750- to 4000-Hz range, called "peak absorbance," and (b) the frequency point at which the maximum occurred, called "peak frequency." The study picked absorbance maxima for all measurements in the training and validation datasets by visual evaluation of individual absorbance graphs and then determined the nearest corresponding numerical absorbance and frequency values in the raw data. A reliability assessment of the analyses between the two students showed an average difference1of 0.0005 in peak absorbance and a difference of 10.18 Hz in peak frequency, indicating excellent reliability.

[0095] In a few instances, absorbance from the training data set exhibited a double-peaked pattern in the 750- to 4000-Hz range, which manifested shallow notches in an otherwise single broad peak in this frequency range. The study observed this odd pattern in 14 measurements from seven ears. Although both peaks (from the double-peaked pattern) were analyzed for these measurements, they were excluded from the computation of normative parameters of the APT. In other instances, in the validation set, 26 absorbance measurements from 26 ears that failed the DPOAE screening did not exhibit distinct peaks in the frequency range of analysis. Instead, absorbance was either flat or gradually sloping. These patterns were consistent with abnormal sound conduction. Therefore, although no peak absorbance or peak frequency could be analyzed, measurements with low and flat absorbance were retained for assessment of APT performance.

[0096] Statistical analysis of system performance. The study performed all statistical analyses using the Statistical Package for the Social Sciences software (Version 19; IBM).

[0097] The study computed the 5th and 95th percentile values on the two absorbance peak variables, peak absorbance, and peak frequency, using data from the training set. These values constituted the parameters for normal absorbance peaks, which were plotted in the form of a rectangle-shaped template that outlined the boundaries for normal absorbance peaks on the absorbance-frequency graph, which can be referred to as the APT. For the computation of normative parameters, the study reduced repeated measurements (i.e., multiple insertions per ear) to one data point per ear. Rather than selecting one measurement at random, the study averaged measurements from repeated insertions within each ear. A matched comparison between Titan and HearlD showed no effect of the system on values of peak absorbance or peak frequency. Therefore, averages within each ear combined repeated measurements irrespective of the system with which they were collected. The resulting APT parameters were based on one average peak absorbance and one average peak frequency per ear (n = 84).

[0098] A separate analysis in the study evaluated the reproducibility of absorbance peak analysis within the training data set, including the 5th and 95th percentile parameters, by splitting the data set. Rather than arbitrarily splitting the data, the study split repeated measurements by system. For this purpose, the study computed within-ear averages separately for HearlD (n - 73) and Titan (n - 82) systems. The study described APT variable distributions (5th, 10th, 25th, 50th, 75th, 90th, and 95th percentile values) from the subset of ears in which data were available from each system (n = 72). Additionally, the study assessed the absorbancepeak data in comparison with the independent validation data set using measurements from ears that passed the DPOAE screening (n = 359), wherein sound conduction was presumed to be normal.

[0099] The resulting APT provided a paradigm for the assessment of normal and abnormal peaks. To evaluate this paradigm, the study used APT to assess absorbance peaks from the validation dataset. Absorbance peaks that fell within the area outlined by the APT were considered normal. Otherwise, abnormal peaks that did not fall within the APT were categorized based on the two assessments: (i) was the value of peak absorbance outside of the 5th-to-95th percentile range? (e.g., was it reduced below the 5th percentile value?), and (ii) was the frequency of the peak shifted outside the 5th-to-95th percentile range, to either lower or higher frequencies? 4 grouping categories resulted from these assessments, including a normal group and three abnormal groups. To further analyze the 4 categories given by the APT, the study compared the proportion of ears that passed or failed the DPOAE screening among these categories. Chi-square analyses determined whether the pass / fail proportions differed among the 4 categories. A significant level of 0.05 (i.e., a=0.05) was Bonferroni-adjusted for repeated pairwise comparisons.

[0100] The study used a receiver operating characteristic (ROC) analysis to evaluate the performance of both peak absorbance and peak frequency in a classification of ears as having "normal" versus "abnormal sound conduction." Using data from the validation set, the study used the pass / fail DPOAE outcomes as the reference standard, where a DPOAE "pass" indicated normal sound conduction and a "fail" indicated abnormal sound conduction. The asymptotic ROC (AROC) analysis summarized test performance, with AROC values close to 0.5 indicating poor distinction between the normal and abnormal groups by the test variables and AROC values close to 1 indicating a perfect distinction (Swets, 1988). The study computed the 95% confidence intervals (Cis) for each AROC value, providing multiple comparisons among the different test variables. The study considered 26 measurements from the DPOAE fail group without a quantifiable absorbance peak. The absence of an absorbance peak was an example of abnormal sound conduction and was included in the test performance analysis. To be included in the statistical analyses, peak frequency values were assigned an arbitrary midpoint frequency value where absorbance was flat or gradually sloping in the 750- to 4000-Hz range. A value of 1675-Hz was assigned for those cases with absent absorbance peaks. Rather than a high- or low-frequency value, the 1675 Hz value was assigned a random point within the parameters of the normal peak frequency range (i.e., 1138-2262 Hz) to avoid biasing thestatistical analysis of peak frequency in either direction. For peak absorbance values, absent peaks were assigned the average value of absorbance in the 750- to 4000-Hz range.

[0101] The study's discriminant function analysis combined peak variables, peak absorbance, and peak frequency into a single multivariate quantity called the discriminant function score (DFS). The peak variables (i.e., independent variables) predicted the DPOAE pass / fail group membership (dependent variable). The discriminant function coefficients for each of the independent variables conveyed their relative importance in the prediction of DPOAE pass / fail outcomes. Also, the study used the resulting coefficients and intercept (constant) values to construct a discriminant function score (DFS) equation that can be used to reproduce the DFS scores based on the raw values of the independent variables. Additionally, the study evaluated the test performance of the multivariate DFS using the ROC analysis in the same manner described herein.

[0102] Absorbance area indices (AAIs) in the study referred to averages of absorbance over specified frequency intervals, wherein the computation of AAI was an adaptation of the reflectance area indices (RAIs) concept that was described by Hunter et al. (2010) and used by others (S. Aithal et al., 2015). The study computed AAIs using absorbance from the validation set over seven frequency intervals: 200-6000, 1000-2000, 1000-4000, 1000-6000, 2000-4000, and 4000-6000 Hz. The study selected these frequency intervals so that AAIs in the study were comparable to RAIs in previous studies. Using ROC analyses, the test performance of these AAIs, indicated by their respective AROC values, was compared to those of absorbance peak variables and the multivariate DFS variable.

[0103] Assessment of sensitivity and specificity of the system. The study computed the sensitivity and specificity rates of the APT using a classification matrix. The study used the normal absorbance peak parameters defined using the training set to classify measurements from the validation data set as either normal or abnormal sound conduction groups. The study used the DPOAE pass / fail outcomes as a reference standard for the actual statuses of normal versus abnormal. Additionally, the study assessed the sensitivity and specificity rates for the "peak absorbance" variable, where normal was defined as peak absorbance > 0.65 (5th percentile value) and abnormal as < 0.65.

[0104] The study also compared the sensitivity and specificity rates of AAI (1000-2000 Hz) and AAI (1000-4000 Hz). The 10th percentile values of the AAIs, computed in the training set, classified measurements from the validation set as belonging to normal versus abnormal sound conduction, where normal was defined as AAI > 10th percentile value and abnormal AAI as < 10th percentile value. The 10th percentile cutoff value was consistent with therecommended RAI values (Hunter et al., 2010). The DPOAE pass / fail outcomes were used as a reference standard for the actual status of normal versus abnormal.

[0105] The study evaluated test-retest with repeated probe reinsertion for peak absorbance and peak frequency using the training set data. The study computed multiple test-retest differences in each ear by subtracting each repeated measurement (e.g., 12, 13, ...) from the first insertion (II). To control for the effect of the measurement system, the study performed these computations separately for Titan and HearlD measurements within the same ear. However, since the resulting test-retest difference values did not show differences when the two systems were compared, the study analyzed all test-retest difference values together in subsequent analyses. The study computed descriptive statistics of test-retest difference values for both absorbance peak variables. In addition, the study computed the absolute values of test-retest differences to describe the upper limit of test-retest for clinical application.

[0106] Experiment Results

[0107] Absorbance peak template (APT) results. The study used peak absorbance and peak frequency values determined from the training set to create the lower and upper boundaries of the APT. Fig. 4A shows a scatter plot of peak absorbance versus peak frequency from 490 analyzed peaks, wherein APT is plotted as a rectangular shape 402. The normative peak absorbance 5th and 95th percentile values (0.65 and 0.99, respectively) are shown by the horizontal boundaries of the rectangle 402, and the peak frequency 5th and 95 th percentile values (1138 and 2262 Hz, respectively) are shown by the vertical boundaries of the rectangle 402. The area inside the template 402 represents the range of normal absorbance peak (peak absorbance and peak frequency) values. The majority of the absorbance peaks occur within the APT (i.e., within the area inside template 402).

[0108] The study assessed the reproducibility of the 5th and 95th percentile values within the training set by splitting the training set by the Titan and HearlD systems used to obtain repeated measurements in each ear. In addition to the 5th and 95th percentile values, Table 1 lists the 10th, 25th, 50th, 75th, and 90th percentile values for peak frequency and peak absorbance (top 4 rows and bottom 4 rows, respectively). In Table 1, values with the training data set were based on averages of within-ear repeated measurements (n = 490 for all ears). Titan and HearlD measurements were repeated within the same ears in 72 out of the 84 ears in the training set.Table 1

[0109] In Table 1 , data from the unsplit training set are presented first, followed by the split Titan and HearlD sets. For peak frequency, the 5th percentile values from the Titan and HearlD sets differed from each other by 76 Hz and differed from the unsplit training set by 9 and 86 Hz, respectively. The 95th percentile peak frequency values differed between the Titan and HearlD sets by 33 Hz and differed from the unsplit training set by 13 Hz and 46 Hz, respectively. Moreover, minimal differences were observed among split and unsplit data sets for any of the peak frequency percentiles (< 90 Hz), indicating consistency in overall data distributions and good reproducibility of peak frequency normal parameters within the training set. For peak absorbance, the 5th percentile values from the Titan and HearlD sets differed from each other by 0.03 and differed from the unsplit training set by 0.04 and 0.02, respectively. The 95th percentile peak absorbance values differed between the Titan and HearlD sets by 0.01 and differed from the unsplit training set by 0.01 and 0.001, respectively. Moreover, minimal differences were observed among split and unsplit data sets for any of the peak absorbance percentiles (< 0.04), indicating consistency in overall data distributions and good reproducibility of peak absorbance normal parameters within the training set.

[0110] The study performed a separate assessment to determine whether APT parameters reproduce in comparison to the independent validation dataset. The study computed the 5th and 95th percentile values in the validation data set and compared them to the values from the training set. The fourth row of Table 1 lists the 5th and 95th percentile values from the validation set for peak frequency, and the last row of Table 1 lists those values for peak absorbance. For peak frequency, the 5th percentile value from the validation set was 1059 Hz,which was 79 Hz (7%) lower than the training set value, and the 95th percentile value was 2670 Hz, which was 408 Hz (18%) greater than the training set value. For peak absorbance, the 5th percentile value was 0.61, which was 0.04 (7%) lower than the training set value, and the 95th percentile value was 0.98, which was 0.01 ( = 1 %) lower than the training set value. In summary, the reproducibility of APT parameters was good, with minimal deviations between the training and validation sets < 7%, except for the 95 th percentile value of peak frequency, which deviated by as 18%.

[0111] Fig. 4B shows a comparison between (i) APT-based assessment of absorbance peaks and (ii) assessment of wideband absorbance across frequency using the traditional normative range. In addition to the determination of normal absorbance peaks, the APT facilitated the assessment of abnormal absorbance peaks based on whether peak absorbance and / or peak frequencies exceed the normative parameters outlined by the APT. To demonstrate this utility, the study used APT to assess absorbance peaks from the validation set, which resulted in 4 categories: (Category A) normal absorbance and frequency, (Category B) normal absorbance and abnormal frequency, (Category C) abnormal absorbance and normal frequency, and (Category D) abnormal absorbance and frequency.

[0112] Fig. 4B shows the 4 categories (e.g., A - D) in subpanels (a) - (d), respectively. In each subpanel, examples of absorbance measurements are plotted across frequencies together with the APT, which is shown by the solid rectangle 402 (shown as 402a - 402d). Absorbance measurements are shown in traces 404 (shown as 404a - 404d) to indicate they were from ears that passed the DPOAE screening and in traces 406 (shown as 406a - 406d) to indicate they were from ears that failed the screening. The overall proportions of DPOAE-Pass (shown as trace 404) versus DPOAE-Fail (shown as trace 406) are shown for each category at the top-left corner of each subpanel. Fig. 4B also includes plots of the absorbance normative (10th to 90th percentile) ranges across frequencies, shown by the gray-shaded regions 408 (shown as 408a - 408d) that are overlayed in the background of each subpanel, to provide a qualitative comparison between the traditional method for absorbance assessment (using the normative absorbance range) and the APT-assessment method in the study.

[0113] In Fig. 4B, subpanel (a) shows 10 example measurements (traces) where the values of peak absorbance and peak frequency fell within the parameters outlined by the APT. These example measurements represented observations from 243 measurements that were grouped in Category A. Absorbance peaks in Category A predicted those ears to have normal sound conduction. By comparison, the use of the traditional normative range shows absorbance valuesin 8 out of the 10 examples were within the (normative range) shaded region 408a across frequency, and two measurements had a small portion of absorbance values below the 10% (lower edge of the shaded region 408a) at frequency intervals between 1700 and 2700 Hz. In these examples, abnormally low absorbance may suggest a marginal increase in impedance due to an increase in stiffness. In summary, measurements classified into Category A, based on their absorbance peaks, also had absorbance values across frequencies within the shaded region 408a, demonstrating an agreement between the two normative methods.

[0114] Subpanel (b) shows 10 example measurements (traces) where peak absorbance values fell within the normal parameters defined by the APT, but peak frequency values did not. Specifically, the heights of absorbance peaks along the vertical axis were between 0.65 (5th percentile value) and 0.99 (95th percentile value), but their frequencies were shifted higher than 2262 Hz (95th percentile) or lower than 1138 Hz (5th percentile). These example measurements represented observations from the 70 measurements that were grouped into Category B. Absorbance peaks with abnormally high frequencies may indicate changes in resonance consistent with increased stiffness (shown by 4 out of the 10 example measurements), whereas peaks exhibiting abnormally low frequencies may indicate changes consistent with the increase in mass loading (shown by 6 out of the 10 example measurements). In comparison to APT, the 4 absorbance measurements with higher-than-normal peak frequencies had values above the shaded normative region 408b at and near frequencies of their peaks (between 2260 and 3000 Hz) and decreased absorbance values below the shaded region 408b at frequencies between 900 and 2100 Hz. These observations indicated an abnormal increase in stiffness and an increased frequency of resonance consistent with the APT-based assessment. For the other 6 measurements, however, assessments using the traditional normative range resulted in more findings. Only 1 of the 6 measurements had absorbance values above the shaded region 408b at the frequency of its peak, whereas the other 5 measurements did not, despite the lower-than-normal shift in peak frequency per APT assessment. Moreover, only 3 of the 6 measurements had absorbance values below the shaded region at frequencies between 1700 and 2700 Hz. Therefore, assessment of absorbance peaks using APT may be more sensitive to abnormalities in sound conduction due to an increase in mass-loading compared to the traditional normative range method.

[0115] A comparison of DPOAE screening outcomes between Category A (in subpanel (a)) and Category B (in subpanel (b)) showed that the proportions of ears that failed the screening increased from 5.8% in ears with normal absorbance peaks (in Category A) to 8.6% in ears with abnormally frequency- shifted absorbance peaks (in Category B). However, thischange in DPOAE pass / fail proportions between the two categories was not statistically significant per chi-square analysis, suggesting a marginal increase in proportions of DPOAE screening fails may be associated with a milder type of abnormality in sound conduction in association with frequency-shifted absorbance peaks in Category B. Examples of absorbance measurements that pass or fail the DPOAE screening with frequency-shifted peaks are shown in subpanel (b) by solid traces 404b and 406b, respectively.

[0116] Subpanel (c) shows 5 example measurements (traces) where peak absorbance values were abnormally low, but peak frequencies were normal. The heights of the principal absorbance peaks along the vertical axis were below 0.65 (5th percentile), but the frequencies of the peaks were between 1138 Hz (5th percentile value) and 2262 Hz (95th percentile value). These example measurements represented 32 measurements that were grouped in Category C. Most absorbance peaks in Category C exhibited a shallow morphology. Assessments using the traditional normative absorbance range method showed that large portions of absorbance values were below the normative shaded regions at frequencies spanning 1-2 octaves within the 1000- to 4000-Hz interval. Both the APT-based assessment of absorbance peaks and the normative range methods indicated abnormal sound conduction.

[0117] Subpanel (d) shows 7 example measurements (traces) where both peak absorbance and peak frequency were abnormal. These example measurements represented 79 measurements that were grouped in Category D. The heights of principal absorbance peaks along the vertical axis were below 0.65 (5th percentile), and the frequencies of the peaks were either lower than 1138 Hz or higher than 2262 Hz. Similar to absorbance peaks in Category C, most absorbance peaks in Category D exhibited a shallow morphology. Also classified under this APT category were absorbance measurements with no discernible absorbance peak (26 out of the 79 measurements). Assessment using the traditional normative absorbance range method showed findings similar to the measurements in subpanel (c), indicating abnormal sound conduction.

[0118] Reduced absorbance peaks in Categories C and D suggested greater abnormality in sound conduction compared to Category B, in which absorbance peaks were discernible and within normal APT-defined parameters for peak absorbance values. This worse sound conduction in Categories C and D was consistent with comparisons of DPOAE pass / fail proportions between Category B and each of Categories C and D. The proportions of ears that failed increased from 8.6% in Category B to 43.8% and 38% in Categories C and D, respectively. Chi-square analyses showed that DPOAE pass / fail proportions differed betweenCategory B and each of Categories C and D, whereas proportions of pass / fail were not significantly different between ears in Categories C and D.

[0119] Test performance results. ROC analyses in the study evaluated the performance of the two absorbance peak test variables (peak absorbance and peak frequency) in the classification of ears as having "normal" versus "abnormal sound conduction" using the DPOAE pass / fail outcomes as a reference standard. For every test variable evaluated, Table 2 shows the AROC values (in the second column) and their 95% confidence interval (CI) values, denoted by lower and upper bound values (in the third and fourth columns, respectively). In Table 2, DPOAE pass / fail outcomes were used as a reference standard for test performance analysis. AROC denotes the areas under the receiver operating characteristic curve, DFS denotes the discriminant function score, AAls denotes the absorbance area indices, DPOAE denotes the distortion-product otoacoustic emission, and RAI denotes the reflectance area index.Table 2

[0120] In Table 2, the AROC for peak absorbance, 0.83 (95% CI [0.77, 0.881), was higher than the AROC for peak frequency, 0.64 (95% CI [0.56, 0.77]). The test performance of peak absorbance and peak frequency was also tested in combination using the discriminant function analysis. The resulting multivariate score (DFS) was directly related to the independent variables using the discriminant function coefficients, as shown in Equation 3.DFS = 6.079 x Peak absorbance + 1.351 x 10 x Peak absorbance - 4.849(Eq. 3)

[0121] In Equation 3, the coefficient values indicated that the contribution of peak frequency (1.351 x 10^) to the DFS was negligible and that peak absorbance (6.079) was the primary contributor to the DFS. Given these outcomes, the AROC for DFS, 0.83 (95% CI [0.77, 0.89]), was similar to the AROC of peak absorbance.

[0122] The instant study compared the test performance of the absorbance peak variables to other methods in previous studies. AAIs were computed similarly to RAIs and at similar frequency intervals that were tested by Hunter et al. (2010). Table 2 shows the AROC values and the 95% Cis for seven AAI variables. For the AAIs with the five highest AROC values, ranging from 0.80 for AAI (200-6000 Hz) to 0.85 for AAI (1000-4000 Hz), the overlap among their 95% Cis indicated the differences in AROC values were insignificant. Moreover, the AROC values of these five AAIs were not significantly different from those of peak absorbance and DFS. The AROC value of AAI (2000-6000 Hz) was higher than the AROC values of AAI (4000-6000 Hz) and peak frequency, and lower than the remaining five AAIs; AROC values of AAI (4000-6000 Hz) and peak frequency were not significantly different from each other.

[0123] Table 2 also shows a comparison to AROC values from Hunter et al. (2010) for RAI over the same frequency intervals. Similar findings for AAI indicated that the five topperforming AAI variables were for those frequency intervals that encompassed the 1000- to 2000-Hz frequency region, and the lowest AROC values were for frequency intervals 2000- 6000 Hz and 4000-6000 Hz. Table 2 also provides AROC values for reflectance at 1500 and 2000 Hz, the highest-performing single-frequency variable reported by Hunter et al. Their AROC values were either similar to or lower than the highest-performing RAI variables in the instant study.

[0124] Results for the sensitivity and specificity of APT. To further describe the performance of APT, the study compared the sensitivity and specificity rates of the APT to those of the two AAIs (1000-2000 Hz and 1000-4000 Hz) with the highest AROC values. In addition, the study included the sensitivity and specificity rates for the peak absorbance variable in this assessment because its AROC value was comparable to the two AAIs (1000- 2000 Hz and 1000- 4000 Hz). Table 3 shows the sensitivity and specificity rates for the four tests, using distortion-product otoacoustic emission pass / fail outcomes as the reference standard.Table 3

[0125] A classification of "abnormal" for APT included cases that were not grouped in the normal Category A. For peak absorbance, "abnormal" was defined as < 0.65 (5th percentile value), and for the AAI variables, "abnormal" was defined as < 10th percentile values, consistent with methods in the previous studies.

[0126] Test-retest difference results. The study computed 298 test-retest difference values for peak frequency and peak absorbance. Table 4 shows the 10th, 25th, 50th, 75th, and 90th percentiles of test-retest difference values and absolute difference values for peak absorbance and peak frequency.Table 4

[0127] Two-tailed (significant value a = .025) single-sample / -tests determined peak absorbance differences were insignificant compared to an absorbance value of 0, t(297) - 1.840, p = .033. Similarly, peak frequency differences were not significantly different from 0 Hz, t(297) = -1.548, p = .061. The median test-retest difference values for peak frequency and peak absorbance were 0.000 and 0.007 Hz, respectively. Table 4 also shows absolute test-retest differences for peak frequency and peak absorbance. The corresponding 90th percentile values, 328.125 Hz for peak frequency and 0.079 Hz for peak absorbance were upper limits of test- retest.

[0128] Discussion

[0129] Discussion #1. The instant study explores the use of a novel normative paradigm that quantifies the mid-frequency absorbance peak in terms of absorbance and frequency variables. Given the link between middle-ear resonance and the mid-frequency peak in newborns and the fact that this peak falls within a frequency interval over which absorbancewas shown to be valuable, the study hypothesized that analysis of absorbance peak may provide means for detection of changes in absorbance as a function of frequency that current normative methods do not provide. Moreover, a normative paradigm based on the characterization of absorbance peaks as a morphological feature may provide a simpler alternative to the current state-of-the-art method for clinical assessment and interpretation. Specifically, the study (a) developed an absorbance peak template (APT) for clinical assessment of outer / middle-ear sound conduction and to evaluate its reliability, (b) described the classification of abnormal categories based on peak absorbance and peak frequency that the APT provides, (c) compared performance of the APT to the traditional normative range in the predication of OAE pass / fail outcomes, and (d) evaluated test-retest differences (with probe reinsertion) in peak absorbance and peak frequency.

[0130] Assessment of absorbance peaks resulted in test performance similar to current state-of-the-art methods for analysis of wideband absorbance measurements. The APT in the instant study provided a simple paradigm for assessing absorbance peaks. Given its discretely defined boundaries for normal measurements and simple categories for assessing abnormal absorbance peaks, APT-based analysis can also be amenable to automation, which is ideal for newborn hearing screening settings. Moreover, APT-based analyses may provide detection of subtle changes in the sound conduction pathway that are otherwise not possible with current normative paradigms. Future characterization of absorbance measurements in ears with abnormal sound conduction pathways using more appropriate reference standard tests may further validate APT methods for abnormal absorbance peaks, which may require using a combination of tests at birth (S. Aithal et al., 2015), or a longitudinal design in combination with a within-subject control (Voss et al., 2016). Further assessment of APT's application may be achieved using longitudinal / retrospective investigations using a combination of frequencyspecific diagnostic tests (e.g., Hunter et al., 2010; Keefe et al., 2003).

[0131] Discussion #2. Universal newborn hearing screening is implemented in 98% of birth hospitals in the United States and worldwide (Gaffney et al., 2014; World Health Organization, 2010). Well-baby newborns receive a hearing screening using an otoacoustic emission (OAE) and / or an automated auditory brainstem response (AABR) test within the first 2-3 days of life. Hearing screening tests are configured to detect hearing loss early on in life so that children with permanent congenital hearing loss may receive timely intervention to prevent long-lasting delays in language, speech, and cognitive development (Joint Committee on Infant Hearing, 2019; Yoshinaga-Itano et al., 1998). However, nonpermanent obstructions due to vemix residue in the outer ear, or residual developmental tissue or fluid in the middle-earcavity, are responsible for 78%-96% of failed screening tests by various estimates (Doyle et al., 2000; Kennedy et al., 2005; Prieve et al., 2000; Thompson et al., 2001; Visscher et al., 2005). This had a negative impact on the referral and follow-up process in the current practice. Because hearing screening workers have no means of discerning the reason for failed tests (i.e., permanent hearing loss vs. temporary conductive hearing loss), most newborns who fail the initial test receive a repeat screening test before discharge and, should they fail again, a referral for additional follow-up testing in outpatient settings. Incorporating a noninvasive immittance test alongside the newborn hearing screening test may enable more targeted rescreens and referrals to follow-up testing, including a timelier referral to diagnostic evaluation and intervention for infants with normal outer / middle ears.

[0132] Wideband acoustic immittance (WAI) refers to a host of measures that convey the response of the outer / middle ear over a wide range of frequencies and are sensitive to peripheral mechanics more generally, including inner ear changes. Because newborns have immature ear canal walls that are not yet ossified, pressurization of the ear canals, such as with tympanometry tests, has been shown to deform the ear canal walls, causing the canal to narrow, collapse with negative pressure, or expand with positive pressure (V. Aithal et al., 2014; Holte et al., 1991). Because WAI testing does not require pressurization of the ear canal, WAI may be used in this population over tympanometry, wherein pressurization is a technical prerequisite for accurate in-ear calibration of admittance units relative to volume units presuming rigid, noncompliant ear canals (Allen, 1986; Lilly & Shanks, 1981). Among the WAI measures, power reflectance and power absorbance have also been used for clinical assessment. Compared to pressure reflectance and pressure absorbance measures, power measures may be advantageous because they are unaffected by standing (pressure) waves and provide more uniform recordings along the depth of the ear canal between the probe tip and the tympanic membrane (TM; Rosowsld et al., 2013). However, variability in power measurements has been noted and attributed to the varying cross-sectional area of the ear canal at different points of its length (Voss et al., 2008, 2013). Power absorbance (absorbance for short) refers to the proportion of acoustic power that is absorbed relative to the total power of the incident stimulus, and power reflectance refers to the proportions that were not absorbed but reflected (from the TM) back to the test probe. A value of 0 absorbance means no sound was absorbed and the entirety of incident power was reflected (reflectance = 1 ), and a value of 1 means that all sounds were absorbed, and none was reflected (reflectance = 0). The cumulative impedance of the middle-ear system determines absorbance / reflectance at the TM. In the instant study, measurements are described in terms of absorbance; reflectance is referredto only when discussing earlier studies. Another advantage of WAI testing is obtaining stable measurements using broadband probe stimuli. While tympanometry testing does not result in stable measurement patterns for probe frequencies > 2000 Hz, advanced techniques for characterization of probe acoustics (i.e., Thevenin equivalent calibration) enable WAI testing at higher probe frequencies, up to 8000 Hz with commercially available systems (Allen, 1986; Keefe et al., 1993; Lilly, 1984; Margolis et al., 1999).

[0133] Evaluating the acoustic response of the outer / middle ear over a wide range of frequencies conveys information about their acoustic mechanics. For example, frequencies at which absorbance values are maximum, which may correspond to minimum impedance values, suggest a resonance frequency. Smaller absorbance values, at lower and higher frequencies than the frequency of maximum absorbance, may be associated with greater stiffness and mass reactance, respectively. The balance between stiffness and mass reactance may result in a shifting of the frequency of resonance, that is, to a higher frequency in a stiffness-dominated system and to a lower frequency in a mass-dominated system (Kei et al., 2016; Rosowsld & Relkin, 2001). This simplified paradigm, which presumes a coherent predominant resonance, has been described as a simple paradigm for assessing and interpreting changes in wideband immittance data (AlMalcadma, Kei, et al., 2023; Withnell et al., 2009). WAI recordings are influenced by compound resonant elements within the middle ear, which may overlap in frequency or appear as distinct coherent maxima or peaks at different frequencies, and the cochlear fluids to which it is coupled (Hudde & Engel, 1998). Nevertheless, the interpretive paradigm of Withnell et al. (2009) is useful when comparing changes in wideband recordings between normal and disordered ears.

[0134] In newborns, wideband absorbance is characterized by a prominent mid-frequency peak between 1000 and 2000 Hz, a smaller low-frequency peak between 300 and 500 Hz, and an additional peak at higher frequencies around 6000 Hz (AlMakadma, Aithal, et al., 2023; Merchant et al., 2010; Voss et al., 2016). A fluid-structure finite-element model by Motallebzadch et al. (2017) describes the resonant elements conveyed by the wideband response in newborns with validated findings in 2- to 4-week-old infants. Their model describes a middle-ear resonance around 1800 Hz, corresponding to the mid-frequency absorbance peak, and a resonance of the cartilaginous ear canal wall around 500 Hz. Their model also predicts a high-frequency middle-ear cavity resonance element around 6000 Hz (Motallebzadeh et al., 2017). The low-frequency absorbance peak has been previously modeled and described in newborns and young infants (Hunter et al., 2015; Keefe et al., 1993; Keefe & Levi, 1996). The frequency range at which the mid-frequency absorbance peak occurs overlapswith regions of high diagnostic performance. Compared to other frequency regions, the performance of reflectance / absorbance measurements in the 1000-2000 Hz range results in the best test performance for assessment of the sound conduction pathway (e.g., S. Aithal et al., 2015; Hunter et al., 2010). (Due to the lack of a gold-standard test of conductive function in newborn screening settings, the pass vs. fail outcome (e.g., on an OAE screening) has been used as a reference standard to indicate normal vs. abnormal sound conduction, respectively, for analysis of test performance.) For example, Hunter et al. (2010) reported areas under the receiver operating characteristic curve (AROC) for reflectance at specific frequencies within this range were greatest at 1500 and 2000 Hz (AROC = 0.88 and 0.9, respectively). When measurements were averaged across frequencies, called reflectance area indices (RAIs), AROCs were at 0.9 for RAI (1000-2000 Hz) and RAI (1000-4000 Hz). A model proposed by Sanlcowsky-Rothe et al. (2022) predicts that WAI measurements in the 1000 to 3000 Hz range are dominated by resonant properties of the eardrum and the middle ear. The sensitivity to abnormal sound conduction in this frequency range may be related to those predicted resonant components (Sankowsky-Rothe et al., 2022), which have been related to the prominent absorbance peak in that range (Motallebzadeh et al., 2017). The instant study investigates the ability of the mid-frequency absorbance peak to detect sound conduction abnormalities. No previous studies investigated the test performance of the absorbance peak in newborns or young infants as a diagnostic feature.

[0135] An aspect of wideband absorbance measurements is the potential to produce pattems / morphological features of absorbance measurements as a function of frequency in association with subtle changes in the acoustic mechanics of the middle ear in the presence of varying degrees and types of dysfunctions (Nakajima et al., 2013). Efforts to characterize and assess frequency-dependent absorbance pattems / features come from adult reports, while such efforts are limited in newborn-related studies. For example, absorbance / reflectance measurements may show a notch at mid-frequencies for some pathologies (Merchant et al., 2015) and a shifting of absorbance peak frequencies to lower or higher frequencies and / or reduction or increase in the size of the peaks in a disorder-dependent manner (Feeney et al., 2003; Nakajima et al., 2012; Voss et al., 2012). Other studies (e.g., Shahnaz et al., 2009) have shown the interaction between absorbance / reflectance and frequency in predicting otosclerosis. Although such frequency-dependent patterns or features are of demonstrable diagnostic value, the current state-of-the-art methods for characterization of normal absorbance measurements compute normative statistics (e.g., lower and upper percentile limits of normal, means / medians) at each frequency point while disregarding frequency-dependent features suchas frequencies of the peaks and slopes. In the newborn-related studies, the ranges of normal, which are indicated by lower limit (e.g., 5th, 10th, or 25th percentile value) and upper limit (e.g., 75th, 90th, or 95th percentile value) values, are plotted at each frequency, producing a normative region (S. Aithal et al., 2017; Hunter et al., 2010; Merchant et al., 2010; Sanford et al., 2009). The ranges of normal (e.g., 10th-90th percentile values) values are large and vary from one frequency to another, between 0.36 and 0.54 (e.g., S. Aithal et al., 2017; Hunter et al., 2010), posing a limitation to the utility of the normative range for clinical interpretation. Disregarding frequency-dependent variability in absorbance measurements, for example, in relationship to distinct morphological features, such as peaks, slopes, and notches, may contribute to these impractically large normal ranges. The current state-of-the-art method for describing absorbance norms does not provide a useful means for assessing frequencydependent measurement features. Also, the ability to discern subtle changes along the frequency variable, for example, small shifts in the frequency of the absorbance peak, may be restricted by the large ranges of normal. Hence, a normative paradigm that considers frequencydependent features of interest may prove advantageous over the current state-of-the-art normative range method.

[0136] Example Computing System

[0137] The exemplary system and method may be implemented (1) as a sequence of computer-implemented acts or program modules running on a computing system and / or (2) as interconnected machine logic circuits or circuit modules within the computing system. The implementation is a matter of choice dependent on the performance and other requirements of the computing system. Accordingly, the logical operations described herein are referred to variously as state operations, acts, or modules. These operations, acts, and / or modules can be implemented in software, in firmware, in special purpose digital logic, in hardware, and any combination thereof. It should also be appreciated that more or fewer operations can be performed than shown in the figures and described herein. These operations can also be performed in a different order than those described herein.

[0138] The computer system is capable of executing the software components described herein for the exemplary method or systems. In an embodiment, the computing device may comprise two or more computers in communication with each other that collaborate to perform a task. For example, but not by way of limitation, an application may be partitioned in such a way as to permit concurrent and / or parallel processing of the instructions of the application. Alternatively, the data processed by the application may be partitioned in such a way as to permit concurrent and / or parallel processing of different portions of a data set by the two ormore computers. In an embodiment, virtualization software may be employed by the computing device to provide the functionality of a number of servers that are not directly bound to the number of computers in the computing device. For example, virtualization software may provide twenty virtual servers on four physical computers. In an embodiment, the functionality disclosed above may be provided by executing the application and / or applications in a cloud computing environment. Cloud computing may comprise providing computing services via a network connection using dynamically scalable computing resources. Cloud computing may be supported, at least in part, by virtualization software. A cloud computing environment may be established by an enterprise and / or can be hired on an as-needed basis from a third-party provider. Some cloud computing environments may comprise cloud computing resources owned and operated by the enterprise as well as cloud computing resources hired and / or leased from a third-party provider.

[0139] In its most basic configuration, a computing device includes at least one processing unit and system memory. Depending on the exact configuration and type of computing device, system memory may be volatile (such as random-access memory (RAM)), non-volatile (such as read-only memory (ROM), flash memory, etc.), or some combination of the two.

[0140] The processing unit may be a standard programmable processor that performs arithmetic and logic operations necessary for the operation of the computing device. While only one processing unit is shown, multiple processors may be present. As used herein, processing unit and processor refers to a physical hardware device that executes encoded instructions for performing functions on inputs and creating outputs, including, for example, but not limited to, microprocessors (MCUs), microcontrollers, graphical processing units (GPUs), and application-specific circuits (ASICs). Thus, while instructions may be discussed as executed by a processor, the instructions may be executed simultaneously, serially, or otherwise executed by one or multiple processors. The computing device may also include a bus or other communication mechanism for communicating information among various components of the computing device.

[0141] Computing devices may have additional features / functionality. For example, the computing device may include additional storage, such as removable storage and nonremovable storage, including, but not limited to, magnetic or optical disks or tapes. Computing devices may also contain network connection(s) that allow the device to communicate with other devices, such as over the communication pathways described herein. The network connection(s) may take the form of modems, modem banks, Ethernet cards, universal serial bus (USB) interface cards, serial interfaces, token ring cards, fiber distributed data interface(FDDI) cards, wireless local area network (WLAN) cards, radio transceiver cards such as code division multiple access (CDMA), global system for mobile communications (GSM), longterm evolution (LTE), worldwide interoperability for microwave access (WiMAX), and / or other air interface protocol radio transceiver cards, and other well-known network devices. Computing devices may also have input device(s) such as keyboards, keypads, switches, dials, mice, trackballs, touch screens, voice recognizers, card readers, paper tape readers, or other well-known input devices. Output device(s) such as printers, video monitors, liquid crystal displays (LCDs), touch screen displays, displays, speakers, etc., may also be included. The additional devices may be connected to the bus in order to facilitate the communication of data among the components of the computing device. All these devices are well-known in the art and need not be discussed at length here.

[0142] The processing unit may be configured to execute program code encoded in tangible, computer-readable media. Tangible, computer-readable media refers to any media that is capable of providing data that causes the computing device (i.e., a machine) to operate in a particular fashion. Various computer-readable media may be utilized to provide instructions to the processing unit for execution. Example tangible, computer-readable media may include but is not limited to volatile media, non-volatile media, removable media, and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. System memory, removable storage, and non-removable storage are all examples of tangible computer storage media.

[0143] Example tangible, computer-readable recording media include but are not limited to, an integrated circuit (e.g., field-programmable gate array or application-specific IC), a hard disk, an optical disk, a magneto-optical disk, a floppy disk, a magnetic tape, a holographic storage medium, a solid-state device, RAM, ROM, electrically erasable program read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices.

[0144] In light of the above, it should be appreciated that many types of physical transformations take place in the computer architecture to store and execute the software components presented herein. It also should be appreciated that the computer architecture may include other types of computing devices, including hand-held computers, embedded computer systems, personal digital assistants, and other types of computing devices known to those skilled in the art.

[0145] In an example implementation, the processing unit may execute program code stored in the system memory. For example, the bus may carry data to the system memory, from which the processing unit receives and executes instructions. The data received by the system memory may optionally be stored on the removable storage or the non-removable storage before or after execution by the processing unit.

[0146] It should be understood that the various techniques described herein may be implemented in connection with hardware or software or, where appropriate, with a combination thereof. Thus, the methods and apparatuses of the presently disclosed subject matter, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine-readable storage medium wherein, when the program code is loaded into and executed by a machine, such as a computing device, the machine becomes an apparatus for practicing the presently disclosed subject matter. In the case of program code execution on programmable computers, the computing device generally includes a processor, a storage medium readable by the processor (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. One or more programs may implement or utilize the processes described in connection with the presently disclosed subject matter, e.g., through the use of an application programming interface (API), reusable controls, or the like. Such programs may be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language, and it may be combined with hardware implementations .

[0147] Although example embodiments of the present disclosure are explained in some instances in detail herein, it is to be understood that other embodiments are contemplated. Accordingly, it is not intended that the present disclosure be limited in its scope to the details of construction and arrangement of components set forth in the following description or illustrated in the drawings. The present disclosure is capable of other embodiments and of being practiced or carried out in various ways.

[0148] In this specification and in the claims that follow, reference will be made to a number of terms, which shall be defined to have the following meanings:

[0149] Throughout the description and claims of this specification, the word “comprise” and other forms of the word, such as “comprising” and “comprises,” means including but not limited to, and are not intended to exclude, for example, other additives, segments, integers, orsteps. Furthermore, it is to be understood that the terms comprise, comprising, and comprises as they relate to various aspects, elements, and features of the disclosed invention also include the more limited aspects of “consisting essentially of” and “consisting of.”

[0150] As used herein, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to an “electrode” includes aspects having two or more such electrodes unless the context clearly indicates otherwise.

[0151] Ranges can be expressed herein as from “about” one particular value and / or to “about” another particular value. When such a range is expressed, another aspect includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by use of the antecedent “about,” it will be understood that the particular value forms another aspect. It should be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint.

[0152] As used herein, the terms “optional” or “optionally” mean that the subsequently described event or circumstance may or may not occur, and that the description includes instances where said event or circumstance occurs and instances where it does not.

[0153] For the terms “for example” and “such as,” and grammatical equivalences thereof, the phrase “and without limitation” is understood to follow unless explicitly stated otherwise.

[0154] The following patents, applications and publications as listed below and throughout this document are hereby incorporated by reference in their entirety herein.Reference list[1] Aithal, S., Kei, J., Aithal, V., Manuel, A., Myers, J., Driscoll, C., & Khan, A. (2017). Normative study of wideband acoustic immittance measures in newborn infants. Journal of Speech, Language, and Hearing Research, 60(5), 1417-1426. https: / / doi.org / 10. 1044 / 2016_JSLHR-H- 16-0237.[2] Aithal, S., Kei, J., Driscoll, C., Khan, A., & Swanston, A. (2015). Wideband absorbance outcomes in newborns: A comparison with high-frequency tympanometry, automated brainstem response, and transient evoked and distortion product Oto-acoustic emissions. Ear and Hearing, 36(5), e237-e250. https: / / doi.org / 10.1097 / AUD.0000000000000175.[3] Aithal, V., Kei, J., Driscoll, C., Swanston, A., Roberts, K., Murakoshi, M., & Wada, H. (2014). Normative sweep fre-quency impedance measures in healthy neonates. Journal of the American Academy of Audiology, 25(04), 343-354. https: / / doi.org / 10.3766 / jaaa.25.4.6.[4] Allen, J. B. (1986). Measurement of eardrum acoustic impedance. In J. B. Allen, J. L. Hall, A. E. Hubbard, S. T. Neely, & A. Tubis (Eds.), Peripheral auditory mechanisms (pp. 44-51). Springer. https: / / doi.org / 10.1007 / 978-3-642-50038-l_6A.[5] AlMakadma, H., Aithal, S., Aithal, V., & Kei, J. (2023). Use of wideband acoustic immittance in neonates and infants. Seminars in Hearing, 44(01), 029-045. https: / / doi.org / 10.1055 / s-0043-1764200.[6] AlMakadma, H., Kei, J., Yeager, D., & Feeney, M. P. (2023). Fundamental concepts for assessment and interpretation of wide-band acoustic immittance measurements. Seminars in Hearing, 44(01), 017-028. https: / / doi.org / 10.1055 / s-0043-1763293.[7] AlMakadma, H., & Prieve, B. A. (2021). Refining measurements of power absorbance in newborns: Probe fit and intrasubject variability. Ear and Hearing, 42(3), 531-546. https: / / doi.org / 10.1097 / AUD.0000000000000954.[8] Demir, E., Afacan, N. N., Celiker, M., Celiker, F. B., Inecikli, M. F., Terzi, S., & Dursun, E. (2019). Can wideband tympa-nometry be used as a screening test for superior semicircular canal dehiscence? Clinical and Experimental Otorhinolaryngol-ogy, 12(3), 249-254. https: / / doi.org / 10.21053 / ceo.2018.01137.[9] Downing, C., Kei, J., & Driscoll, C. (2022). Measuring resonance frequency of the middle ear in school-aged children: Potential applications for detecting middle ear dysfunction. Intema-tional Journal of Audiology, 62(11), 1076-1083. https: / / doi.org / 10.1080 / 14992027.2022.2135033.

[0010] Doyle, K. J., Rodgers, P., Fujikawa, S., & Newman, E. (2000). External and middle ear effects on infant hearing screening test results. Otolaryngology — Head & Neck Surgery, 122(4), 477-481.

[0011] Feeney, M. P., Grant, I. L., & Marryott, L. P. (2003). Wideband energy reflectance measurements in adults with middle-ear dis-orders. Journal of Speech, Language, and Hearing Research, 46(4), 901-911. https: / / doi.org / 10. 1044 / 1092-4388(2003 / 070).

[0012] Gaffney, M., Eichwald, J., Gaffney, C., & Alam, S. (2014). Early hearing detection and intervention among infants — Hearing screen-ing and follow-up survey, United States, 2005- 2006 and 2009- 2010. Morbidity and Mortality Weekly Report, 63(2), 20-26.

[0013] Holte, L., Margolish, R. H., & Cavanaugh, R. M. (1991). Devel-opmental changes in multifrequency tympanograms. Interna-tional Journal of Audiology, 30(1), 1-24. https: / / doi.org / 10. 3109 / 00206099109072866.

[0014] Hudde, H., & Engel, A. (1998). Measuring and modeling basic properties of the human middle ear and ear canal. Part I: Model structure and measuring techniques. Acta Acustica United with Acustica, 84(4), 720-738.

[0015] Hunter, L. L„ Feeney, M. P„ Miller, J. A. L„ Jeng, P. S., & Bohning, S. (2010). Wideband reflectance in newborns: Nor-mative regions and relationship to hearing screening results. Ear and Hearing, 31(5), 599-610. https: / / doi.org / 10.1097 / AUD.0b013e3181e40ca7.

[0016] Hunter, L. L., Keefe, D. H., Feeney, M. P., Fitzpatrick, D. F., & Lin, L. (2015). Longitudinal development of wideband reflec-tance tympanometry in normal and at-risk infants. Hearing Research, 340, 3-14. https: / / doi.org / 10.1016 / j-heares.2015.12.014.

[0017] Hunter, L. L., & Shahnaz, N. (2014). Physical principles of aural acoustic Immittance. In J. W. Hall III & V. Ramachandran (Eds.), Acoustic immittance measures: Basic and advanced practice (pp. 23-40). Plural Publishing.

[0018] Johansen, H. (1948). Relation of audiograms to the impedance formula. Acta Oto- Laryngologica, 36(Suppl. 74), 65-75. https: / / doi.org / 10.3109 / 00016484809131988

[0019] Joint Committee on Infant Hearing. (2019). Year 2019 position statement: Principles and guidelines for early hearing detection and intervention programs. Journal of Early Hearing Detection and Intervention, 4(2), 1-44. https: / / doi.org / 10.15142 / fptk-b748.

[0020] Keefe, D. H., Bulen, J. C., Arehart, K. H., & Burns, E. M. (1993). Ear-canal impedance and reflection coefficient in human infants and adults. The Journal of the Acoustical Society of America, 94(5), 2617-2638. https: / / doi.org / 10.1121 / L407347.

[0021] Keefe, D. H„ Gorga, M. P„ Neely, S. T„ Zhao, F„ & Vohr, B. R. (2003). Ear-canal acoustic admittance and reflectance measurements in human neonates. II. Predictions of middle-ear dysfunction and sensorineural hearing loss. The Journal of the Acoustical Society of America, 113(1), 407-422. https: / / doi. org / 10.1121 / 1.1523388.

[0022] Keefe, D. H., & Levi, E. (1996). Maturation of the middle and external ears: Acoustic power-based responses and reflectance tympanometry. Ear and Hearing, 17(5), 361-373. https: / / doi. org / 10. 1097 / 00003446-199610000-00002.

[0023] Kei, J., Aithal, V., Aithal, S., Anderson, S., & Wright, D. (2016). Predicting the characteristics of wideband absorbance in ears with middle-ear pathologies. The Journal of the Acoustical Society of America, 140(Suppl. 4), 3263-3263. https: / / doi.org / 10.1121 / 1.4970341.

[0024] Kei, J., Sanford, C. A., Prieve, B. A., & Hunter, L. L. (2013). Wideband acoustic immittance measures: Developmental characteristics (0 to 12 months). Ear and Hearing, 34(Suppl. 1), 17s-26s. https: / / doi.org / 10.1097 / AUD.0b013e31829db914.

[0025] Kennedy, C., McCann, D., Campbell, M. J., Kimm, L., & Thornton, R. (2005). Universal newborn screening for perma-nent childhood hearing impairment: An 8-year follow-up of a controlled trial. The Lancet, 366(9486), 660-662. https: / / doi. org / 10.1016 / S0140-6736(05)67138-3.

[0026] Lilly, D., & Shanks, J. (1981). Acoustic immittance of an enclosed volume of air. In G. Popelka (Ed.), Hearing assess-ment with the acoustic reflex (pp. 145-160). Grune & Stratton.

[0027] Lilly, D. J. (1984). Multiple frequency, multiple component tym-panometry: New approaches to an old diagnostic problem. Ear and Hearing, 5(5), 300-308. https : / / doi.org / l 0.1097 / 00003446- 198409000-00007.

[0028] Margolis, R. H., Saly, G. L., & Keefe, D. H. (1999). Wideband reflectance tympanometry in normal adults. The Journal of the Acoustical Society of America, 106(1), 265-280. https: / / doi. org / 10. 1121 / 1.427055.

[0029] Masud, S. F. (2020). Diagnosis of mechanical ear pathologies using structure-based modeling and machine learning tech-niques. [Doctoral dissertation]. Harvard University.

[0030] Merchant, G. R., Horton, N. J., & Voss, S. E. (2010). Normative reflectance and transmittance measurements on healthy newborn and 1 -month-old infants. Ear and Hearing, 31(6), 746- 754. https: / / doi.org / 10.1097 / AUD.0b013e3181e68e68.

[0031] Merchant, G. R., Roosli, C., Niesten, M. E., Hamade, M. A., Lee, D. J., McKinnon, M. L., Ulku, C. H., Rosowski, J. J., Merchant, S. N., & Nakajima, H. H. (2015). Power reflectance as a screening tool for the diagnosis of superior semicir-cular canal dehiscence. Otology and Neurotology, 36(1), 172- 177. https: / / doi.org / 10.1097 / MAO.0000000000000294.

[0032] Motallebzadeh, H„ Maftoon, N„ Pitaro, J., Funnell, W. R. J., & Daniel, S. J. (2017). Fluid-structure finite-element modelling and clinical measurement of the wideband acoustic input admittance of the newborn ear canal and middle ear. Journal of the Association for Research in Otolaryngology, 18(5), 671- 686. https: / / doi.org / 10.1007 / sl0162-017-0630-z

[0033] Nakajima, H. H., Pisano, D. V., Roosli, C., Hamade, M. A., Merchant, G. R., Mahfoud, L., Halpin, C. F., Rosowski, J. J., & Merchant, S. N. (2012). Comparison of ear-canal reflectance and umbo velocity in patients with conductive hearing loss: A preliminary study. Ear and Hearing, 33(1), 35-43. https: / / doi.org / 10.1097 / AUD.0b013e31822ccba0.

[0034] Nakajima, H. H., Rosowski, J. J., Shahnaz, N., & Voss, S. E. (2013). Assessment of ear disorders using power reflectance. Ear and Hearing, 34(Suppl. 1), 48s-53s. https: / / doi.org / 10. 1097 / AUD.0b013e31829c964d.

[0035] Prieve, B. A., Dalzell, L., Berg, A., Bradley, M., Cacace, A., Campbell, D., DeCristofaro, J., Gravel, J., Greenberg, E., & Gross, S. (2000). The New York state universalnewborn hearing screening demonstration project: Outpatient outcome measures. Ear and Hearing, 21(2), 104-117. https: / / doi.org / 10. 1097 / 00003446-200004000-00005.

[0036] Rosen, B., AlMakadma, H., & Sanford, C. A. (2020, March 5-7). An absorbance peak template for assessment of newborn con-ductive pathways [Poster presentation]. 47th Annual Scientific and Technology Conference of the American Auditory Society, Scottsdale, AZ, United States, https: / / aas.memberclicks.net / assets / 19_Posters.pdf.

[0037] Rosowski, J. J., Nakajima, H. H., Hamade, M. A., Mafoud, L., Merchant, G. R., Halpin, C. F., & Merchant, S. N. (2012). Ear-canal reflectance, umbo velocity and tympanometry in normal hearing adults. Ear and Hearing, 33(1), 19-34. https: / / doi.org / 10.1097 / AUD.0b013e31822ccb76.

[0038] Rosowski, J. J., & Relkin, E. M. (2001). Introduction to the anal-ysis of the middle ear function. In A. F. Jahn & J. Santos-Sacchi (Eds.), Physiology of the ear (pp. 161-189). Singular / Thomson Learning. http: / / books.google.com / books?id=giFuKH5u7O0C.

[0039] Rosowski, J. J., Stenfelt, S., & Lilly, D. (2013). An overview of wideband Immittance measurements techniques and terminology: You say absorbance, I say reflectance. Ear and Hearing, 34(Suppl. 1), 9s-16s. https: / / doi.org / 10.1097 / AUD. 0b013e31829d5al4.

[0040] Sanford, C. A., Keefe, D. H., Liu, Y.-W., Fitzpatrick, D., McCreery, R. W., Lewis, D. E., & Gorga, M. P. (2009). Sound-conduction effects on distortion-product Otoacoustic emission screening outcomes in newborn infants: Test perfor-mance of wideband acoustic transfer functions and 1-kHz tympanometry. Ear and Hearing, 30(6), 635-652. https: / / doi. org / 10. 1097 / AUD.0b013e3181 b61 cdc.

[0041] Sankowsky-Rothe, T., van de Par, S., & Blau, M. (2022). Para-metric model of young infants’ eardrum and ear canal imped-ances supporting immittance measurement results. Part I: Development of the model. Acustica, 6, 53. https: / / doi.org / 10. 1051 / aacus / 2022047.

[0042] Shahnaz, N., Bork, K., Polka, L., Longridge, N., Bell, D., & Westerberg, B. D. (2009). Energy reflectance and tympanome-try in normal and otosclerotic ears. Ear and Hearing, 30(2), 219-233. https: / / doi.org / 10.1097 / AUD.0b013e3181976al4.

[0043] Swets, J. A. (1988). Measuring the accuracy of diagnostic sys-tems. Science, 240(4857), 1285-1293. https: / / doi.org / 10.1126 / science.3287615.

[0044] Thompson, D. C., McPhillips, H., Davis, R. L., Lieu, T. A., Homer, C. J., & Helfand, M. (2001). Universal newborn hearing screening: Summary of evidence. Journal of the American Medical Association, 286(16), 2000-2010. https: / / doi.org / 10.100 l / jama.286.16.2000.

[0045] Visscher, M. O., Narendran, V., Pickens, W. L., LaRuffa, A. A., Meinzen-Derr, J., Allen, K., & Hoath, S. B. (2005). Vemix caseosa in neonatal adaptation. Journal of Perinatology, 25(7), 440-446. https: / / doi.org / 10.1038 / sj.jp.7211305.

[0046] Voss, S. E., Herrmann, B. S., Horton, N. J., Amadei, E. A., & Kujawa, S. G. (2016). Reflectance measures from infant ears with normal hearing and transient conductive hearing loss. Ear and Hearing, 37(5), 560-571. https: / / doi.org / 10.1097 / AUD.0000000000000293.

[0047] Voss, S. E., Horton, N. J., Woodbury, R. R., & Sheffield, K. N. (2008). Sources of variability in reflectance measurements on normal cadaver ears. Ear and Hearing, 29(4), 651 — 665. https: / / doi.org / 10.1097 / AUD.0b013e318174f07c.

[0048] Voss, S. E., Merchant, G. R., & Horton, N. J. (2012). Effects of middle-ear disorders on power reflectance measured in cadav-eric ear canals. Ear and Hearing, 33(2), 195-208. https: / / doi. org / 10.1097 / AUD.0b013e31823235b5

[0049] Voss, S. E., Stenfelt, S., Neely, S. T., & Rosowski, J. J. (2013). Factors that introduce intrasubject variability into ear-canal absorbance measurements. Ear and Hearing, 34(Supp. 1), 60s- 64s. https: / / doi.org / 10.1097 / AUD.0b013e31829cfd64.1501 Withnell, R. H., Parent, P., Jeng, P. S., & Allen, J. B. (2009). Using wideband reflectance to measure the impedance of the middle ear. The Hearing Journal, 62(10), 36-38. https: / / doi. org / 10.1097 / 01.HJ.0000361848.81466.97.

[0051] World Health Organization. (2010). Newborn and infant hearing screening: Current issues and guiding principles for action. WHO Press. RetrievedAugust5, 2023, from http: / / www.who.int / blindness / publications / Newborn_and_Infant_Hearing_Screening_Report. pdf

[0052] Yoshinaga-Itano, C., Sedey, A. L., Coulter, D. K., & Mehl, A. L. (1998). Language of early-and later-identified children with hearing loss. Pediatrics, 102(5), 1161-1171. https: / / doi.org / 10. 1542 / peds.102.5.1161.

[0053] AlMakadma, H. A., Reed, B. F., & Sanford, C. A. (2023a). An Absorbance Peak Template for Clinical Assessment of Sound Conduction in Newborn Ears. Journal ofSpeech, Language, and Hearing Research, 66(12), 5152-5168.

[0054] AlMakadma, H., Aithal, S., Aithal, V., & Kei, J. (2023b). Use of wideband acoustic immittance in neonates and infants. Seminars in Hearing, 44(01), 029-045.

Claims

What is claimed is:

1. A system for screening abnormal sound conduction in a newborn or infant subject, the system comprising: a processor; and a memory having instructions stored thereon, wherein when executed, the instructions cause the processor to: receive wideband acoustic immittance (WAI) data comprising a wideband absorbance (WBA) measurement derived from reflectance responses recorded from an auditory canal of a subject in response to an application of a plurality of interrogative signals to the auditory channel by a piece of equipment comprising a vibration generation source, wherein the plurality of interrogative signals includes signals at different primary frequency components in a broadband signal; determine a value for mid- frequency absorbance peak defined as a maximum absorbance in a 750 Hz to 4000 Hz range and a corresponding frequency of the maximum absorbance for a portion of the plurality of interrogative signals; and generate, via a display, (i) an absorbance frequency plot over a wide range of frequencies, including the determined value for the mid-frequency absorbance peak from the measurement and (ii) a graphical element comprising an absorbance peak template for the mid-frequency absorbance peak, wherein presence of a determined value for the mid-frequency absorbance peak in a region outside of the graphical element corresponding to the absorbance peak template and / or absence of a determined value for the mid- frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of the determined value for the mid- frequency absorbance peak in a region being inside the graphical element corresponding to the absorbance peak template indicates a normal sound conduction condition.

2. The system of claim 1, wherein the instructions further cause the processor to: apply a windowing function to one or more portions of the WAI data; and determine a value for a mid- frequency absorbance peak defined as a maximum absorbance in each windowed portion of the WAI data and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

3. The system of any one of claims 1-2, wherein the instructions further cause the processor to: determine a second-derivative function of the WAI data; and determine, using the second-derivative function, a value for mid- frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz range and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

4. The system of any one of claims 1-3, wherein the WAI data further comprises an impedance magnitude and an impedance phase; and wherein the instructions further cause the processor to: determine a value for an impedance minimum in the 750 Hz to 4000 Hz and a corresponding frequency of the impedance minimum for each of a portion of the plurality of interrogative signals; determine a frequency where the impedance phase is zero for a portion of the plurality of interrogative signals; and determine a value for mid- frequency absorbance peak defined as a maximum absorbance and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

5. The system of any one of claims 1-4, wherein, when presence of a determined value for the mid-frequency absorbance peak is in a region outside of the graphical element corresponding to the absorbance peak template or when a determined value for the midfrequency absorbance peak is absent, the instructions further cause the processor to: determine a classification of an abnormal sound conduction condition for the subject based on the location or absence of the mid-frequency absorbance peak.

6. The system of claim 5, wherein, when the presence of a determined value for the midfrequency absorbance peak is in a region outside of the graphical element corresponding to the absorbance peak template or when a determined value for the mid-frequency absorbance peak is absent, the instructions further cause the processor to: determine, using a trained machine learning algorithm, a classification of an abnormal sound conduction condition for the subject based on the location or absence of the mid-frequency absorbance peak, wherein the trained machine learning algorithm employed atraining set annotated to include at least one of (i) a status of acoustic leaks and ear canal tip fit, (ii) a status of collapsing ear canal walls (e.g., suspected, collapsed, open), (iii) status of noise or incomplete sequence of WB A testing, or (iv) a combination thereof.

7. The system of any one of claims 1-6, wherein the instructions further cause the processor to: determine, using a trained machine learning algorithm, a value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

8. The system of any one of claims 1-7, wherein the system further comprises a vibration generation source and a vibration recorder.

9. The system of any one of claims 1-8, wherein the graphical element comprises two or more absorbance peak templates; and wherein presence of a determined value for the mid-frequency absorbance peak in a region outside of the graphical element corresponding to at least one of the absorbance peak templates or absence of a determined value for the mid-frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of all of the determined value for the mid- frequency absorbance peak in a region being inside the graphical element corresponding to all absorbance peak templates indicates a normal sound conduction condition.

10. The system of any one of claims 1-9, wherein the system is configured as an otoacoustic emissions (OAE) device.

11. A method for screening abnormal sound conduction in a newborn or infant subject, the method comprising: a) receiving wideband acoustic immittance (WAI) data comprising a wideband absorbance (WBA) measurement derived from reflectance responses recorded from an auditory canal of a subject in response to an application of a plurality of interrogative signals to the auditory channel by a piece of equipment comprising a vibration generation source,wherein the plurality of interrogative signals includes signals at different primary frequency components in a broadband signal that includes a wide range of frequency components; b) determining a value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz and a corresponding frequency of the maximum absorbance for a portion of the plurality of interrogative signals; and c) generating, via a display, (i) an absorbance frequency plot over a wide range of frequencies, including the determined value for the mid-frequency absorbance peak from the measurement and (ii) a graphical element comprising an absorbance peak template for the mid- frequency absorbance peak, wherein presence of a determined value for the midfrequency absorbance peaks in a region outside of the graphical element corresponding to the absorbance peak template or absence of a determined value for the mid- frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of the determined value for the mid-frequency absorbance peak in a region being inside the graphical element corresponding to the absorbance peak template indicates a normal sound conduction condition.

12. The method of claim 11 , wherein step b) further comprises: applying a windowing function to one or more portions of the WAI data; determining a value for a mid- frequency absorbance peak defined as a maximum absorbance in each windowed portion of the WAI data and a corresponding frequency of the maximum absorbance for each for a portion of the plurality of interrogative signals.

13. The method of any one of claims 11-12, wherein step b) further comprises: determining a second-derivative function of the WAI data; and determining, using the second-derivative function, a value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

14. The method of any one of claims 11-13, wherein the WAI data further comprises an impedance magnitude and an impedance phase, and wherein step b) further comprises:determining a value for an impedance minimum in the 750 Hz to 4000 Hz and a corresponding frequency of the impedance minimum for each of a portion of the plurality of interrogative signals; determining a frequency where the impedance phase is zero for a portion of the plurality of interrogative signals; and determining a value for mid-frequency absorbance peak defined as a maximum absorbance and a corresponding frequency of the maximum absorbance for each of a portion of the plurality of interrogative signals.

15. The method of any one of claims 11-14, wherein step b) further comprises: determining, when the presence of a determined value for the mid-frequency absorbance peak is in a region outside of the graphical element corresponding to the absorbance peak template or when a determined value for the mid-frequency absorbance peak is absent, a classification of an abnormal sound conduction condition for the subject based on the location or absence of the mid-frequency absorbance peak.

16. The method of claim 15, wherein step b) further comprises: determining, when the presence of a determined value for the mid-frequency absorbance peak is in a region outside of the graphical element corresponding to the absorbance peak template or when a determined value for the mid-frequency absorbance peak is absent, using a trained machine learning algorithm, a classification of an abnormal sound conduction condition for the subject based on the location or absence of the mid-frequency absorbance peak.

17. The method of any one of claims 11-16, wherein step b) is executed using a trained machine learning algorithm.

18. The method of any one of claims 11-17, wherein the graphical element comprises two or more absorbance peak templates, and wherein presence of a determined value for the mid-frequency absorbance peak in a region outside of the graphical element corresponding to at least one of the absorbance peak templates or absence of a determined value for the mid-frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of all of the determined value for the mid- frequency absorbance peak in a region being inside thegraphical element corresponding to all absorbance peak templates indicates a normal sound conduction condition.

19. A method for using wideband absorbance in a hearing screening (NHS) procedure for a newborn or infant subject, the method comprising: receiving a first wideband acoustic immittance (WAI) data comprising a wideband absorbance (WBA) measurement derived from reflectance responses recorded from an auditory canal of the subject in response to an application of a plurality of interrogative signals to an auditory channel by a piece of equipment comprising a vibration generation source, wherein the plurality of interrogative signals includes signals at different primary frequency components in a broadband signal that includes a wide range of frequency components; and in response to the first WAI data failing a first test performed by an otoacoustic emissions (OAE) device or an automated auditory brainstem response (AABR) device: determining a first value for mid-frequency absorbance peak defined as a maximum absorbance in a 750 Hz to 4000 Hz range and a first corresponding frequency of the maximum absorbance for a portion of the plurality of interrogative signals, and generating, via a display, (i) a first absorbance frequency plot over a wide range of frequencies, including the first determined value for the mid-frequency absorbance peak from the measurement and (ii) a graphical element comprising an absorbance peak template for the mid-frequency absorbance peak, wherein presence of a first determined value for the mid- frequency absorbance peaks in a region outside of the graphical element corresponding to the absorbance peak template or absence of a determined value for the mid-frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of the first determined value for the mid- frequency absorbance peak in a region being inside the graphical element corresponding to the absorbance peak template indicates a normal sound conduction condition.

20. The method of claim 19 further comprising: in response to the subject being determined as having the normal sound conduction condition, outputting the first absorbance frequency plot and / or the first determined value forthe mid-frequency absorbance peak as a first data object, wherein the outputted first data object is subsequently employed in screening or diagnostics devices.

21. The method of claim 19 further comprising: in response to the subject being determined as having the abnormal sound conduction condition: receiving a second wideband acoustic immittance (WAI) data derived from reflectance responses recorded from the auditory canal of the subject in response to the application of the plurality of interrogative signals to the auditory channel by the piece of equipment comprising the vibration generation source, wherein the plurality of interrogative signals includes signals at different primary frequency components in the broadband signal that includes the wide range of frequency components, and in response to the second WAI data failing a second test performed by an otoacoustic emissions (OAE) device or an automated auditory brainstem response (AABR) device: determining a second value for mid-frequency absorbance peak defined as a maximum absorbance in the 750 Hz to 4000 Hz range and a second corresponding frequency of the maximum absorbance for a portion of the plurality of interrogative signals, and generating, via a display, a second absorbance frequency plot over a wide range of frequencies, including the second determined value for the midfrequency absorbance peak from the measurement, wherein presence of a second determined value for the mid- frequency absorbance peaks in a region outside of the graphical element corresponding to the absorbance peak template or absence of a determined value for the mid- frequency absorbance peak indicates an abnormal sound conduction condition for the subject, and wherein presence of the second determined value for the mid- frequency absorbance peak in a region being inside the graphical element corresponding to the absorbance peak template indicates a normal sound conduction condition.

22. The method of claim 19 further comprising: in response to the first WAI data passing the first test performed by the otoacoustic emissions (OAE) device or the automated auditory brainstem response (AABR) device:determining one or more risk factors relating to the hearing of the subject; and outputting one or more risk factors as a second data object, wherein the outputted second data object is subsequently employed in screening or diagnostic devices.

23. The method of claim 21 further comprising: in response to the subject being determined as having the normal sound conduction condition after the second absorbance frequency plot is generated, outputting the second absorbance frequency plot and / or the second determined value for the mid-frequency absorbance peak as a third data object, wherein the outputted third data object is subsequently employed in screening or diagnostics devices.

24. The method of claim 21 further comprising: in response to the subject being determined as having the abnormal sound conduction condition after the second absorbance frequency plot is generated, outputting the second absorbance frequency plot and / or the second determined value for the mid-frequency absorbance peak as a fourth data object, wherein the outputted fourth data object is subsequently employed for an outpatient rescreening procedure.

Images