Methods and systems for screening auditory health of a subject

Abstract

The auditory health of a subject is assessed by providing one or more audio stimuli to the subject, receiving one or more responses from the subject in response to the stimuli, and determining a heari

Description

BACKGROUND

[0001] Hearing loss refers to a reduction or impairment in a person’s ability to hear sounds compared to what is considered normal. Hearing loss can affect people of all ages and can be caused by various factors, including genetics, aging, noise exposure, medical conditions, and more. Hearing loss can be categorized into different types such as conductive, sensorineural, and mixed hearing loss. Hearing loss can also be categorized by degrees, such as mild, moderate, severe, or profound, based on the affected part of the auditory system and the severity of impairment. Hearing loss can have a profound impact on an individual’s life. Therefore, early detection and frequent monitoring is key to providing subject specific treatments.

[0002] Currently, hearing loss evaluations are performed in specialized clinics or settings with sound isolation rooms and highly sensitive hearing assessment equipment. Thus, to detect hearing loss at an early stage in a widely accessible manner, there is an unmet need for methods and systems capable of measuring a subject’s auditory health at the point of need or point of care without requiring specialized equipment and available attending or physician level healthcare providers. SUMMARY

[0003] Aspects of the invention are set out in the independent claims and preferred features are set out in the dependent claims. Features associated with one aspect may be applied to other aspects alone or in combination. The present disclosure provides a method of providing an auditory health recommendation to a subject, comprising: providing one or more audio stimuli to a subject’s ear with a computing device; receiving one or more responses from the subject in response to the one or more audio stimuli with the computing device; determining a hearing classification of the subject from the one or more responses; and providing the auditory health recommendation to the subject based on the hearing classification. In some embodiments, the auditory health recommendation comprises a referral to a health care professional. In some embodiments, the auditory health recommendation is provided to the subject on the computing device, a different personal computing device, website portal, personal computer, or any combination thereof. In some embodiments, the one or more audio stimuli are provided to the subject’s ear through a headphone or ear bud. In some embodiments, the headphone or the ear bud are wired or wireless. In some embodiments, determining the hearing classification of the subject comprises providing the one or more responses of the subject as an input to one or more trained machine learning algorithms or one or more predictive models that output the hearing classification of the subject. In some embodiments, the one or more trained machine learning algorithms or the one or more predictive models are trained with one or more responses of a plurality of subjects in response to one or more audio stimuli and a corresponding hearing classification of the plurality of subjects. In some embodiments, the one or more audio stimuli comprise audio stimuli of one or more frequencies and / or one or more intensities. Each audio stimulus may have an associated frequency and intensity. In some embodiments, an audio stimulus of the one or more audio stimuli is provided to the subject at a frequency of the one or more frequencies with at least 20 occurrences, and wherein each occurrence of the at least 20 occurrences comprises a different intensity of the audio stimulus. In some embodiments, the one or more frequencies comprise up to about 4 frequencies (e.g. 1, 2, 3 or 4 frequencies) or up to about 8 frequencies (e.g. 5, 6, 7, or 8 frequencies). In some embodiments, the one or more frequencies comprise: about 250 Hz, about 500 Hz, about 1000 Hz, about 2000 Hz, about 3000 Hz, about 4000 Hz, about 6000Hz, about 8000Hz, one or more frequencies in the range 250 Hz to 8000 Hz, or any combination thereof. In some embodiments, the one or more frequencies comprise one or more frequency bands. In some embodiments, the one or more frequency bands comprise frequency ranges of about 100Hz to about 999Hz, about 1000Hz to about 3999Hz, about 4000Hz to about 8000Hz, or any combination thereof. The one or more frequency bands may comprise frequencies of at least around 100 Hz, at least around 1000 Hz or at least around 4000 Hz, and / or the one or more frequency bands may comprise frequencies of no greater than around 999 Hz, no greater than around 3999 Hz or no greater than around 8000 Hz. In some embodiments, the subject’s hearing classification is determined from the subject’s one or more responses to at least 4 frequencies of the one or more frequencies of the one or more audio stimuli. In some embodiments, the subject’s hearing classification is determined from an average of the subject’s one or more responses from at about least 2 frequency bands, at least about 3 frequency bands, or at least 4 frequency bands of the one or more frequencies of the one or more audio stimuli. In some embodiments, the one or more responses from the subject comprise an intensity of the one or more audio stimuli at (or above) which the subject provided a response of detecting the one or more audio stimuli, which may also be referred to as hearing thresholds. In some embodiments, the subject’s hearing classification comprises normal hearing when their hearing thresholds are within normal limits or a normal range, or limits associated with no hearing loss. For example, the subject’s hearing classification may comprise normal hearing when the intensity (at which the one or more stimuli are detected) comprises up to about 20dB, up to about 25dB, or up to about 40dB. In some embodiments, the hearing classification of the subject comprises normal hearing when the intensity of the one or more audio stimuli comprises about -lOdB to about 25dB. In some embodiments, the hearing classification of the subject comprises mild hearing loss when the intensity of the one or more audio stimuli comprises at least about 21 dB, at least about 26 dB or at least about 40 dB, and / or no more than about 26 dB or no more than about 49 dB, for example about 26dB to about 39dB. In some embodiments, the hearing classification of the subject comprises moderate hearing loss when the intensity of the one or more audio stimuli comprises at least about 26 dB, at least about 40 dB or at least about 50 dB and / or no more than about 59 dB, no more than about 69 dB or no more than about 79 dB, for example about 40dB to about 69dB. In some embodiments, the hearing classification of the subject comprises severe hearing loss when the intensity of the one or more audio stimuli comprises at least about 60 dB, at least about 70 dB or at least about 80 dB, and / or no more than about 74 dB, no more than about 84 dB, no more than about 94 dB or no more than about 104 dB, for example about 70dB to about 94dB. In some embodiments, the hearing classification of the subject comprises profound hearing loss when the intensity of the one or more audio stimuli comprises at least about 85 dB, at least about 95 dB or at least about 105 dB, and / or no more than about 100 dB, no more than about 110 dB, no more than about 120 dB or no more than about 130 dB, for example about 95dB to about 120dB, or at least around 95 dB. The hearing classification of the subject may comprise hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of at least around 26 dB, at least around 40 dB, at least around 70 dB and / or at least around 95 dB. In some embodiments, determining the hearing classification comprises determining whether the subject has one or more of: normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, and / or an unfamiliar hearing profile. In some embodiments, the subject’s ear comprises a first ear and a second ear, wherein the one or more responses comprise a first set of one or more responses from the one or more audio stimuli provided to the first ear and a second set of one or more responses from providing the one or more audio stimuli to the second ear. In some embodiments, the subject’s hearing classification comprises asymmetric hearing when an intensity of at least two responses of the one or more responses of the subject’s first ear differ by at least about 15dB, or at least about 20dB from an intensity of at least two responses from the one or more responses of the subject’s second ear. For example, the subject’s hearing classification may comprise asymmetric hearing when, for each of two or more frequencies, intensity thresholds of the responses at the respective frequency differ by at least 15 dB between the first and second ears of the subject. In other words, the hearing classification may comprise asymmetric hearing when the first set of responses for the subject’s first ear comprises a first intensity threshold (e.g. minimum detected intensity of audio stimuli) at a first frequency and a second intensity threshold at a second frequency, and the second set of responses for the subject’s second ear comprises a first intensity threshold at the first frequency and a second intensity threshold at the second frequency, wherein the first intensity thresholds of the first and second ears at the first frequency differ by at least about 15 dB or at least about 20 dB, and the second intensity thresholds of the first and second ears at the second frequency differ by at least about 15 dB or at least about 20 dB. In some embodiments, determining the hearing classification of the subject comprises comparing the one or more responses of the subject to a library of one or more responses to the one or more audio stimuli associated with one or more hearing classifications. In some embodiments, the period of time comprises at least about 2 seconds or at least about 3 seconds. In some embodiments, the recommendation of the subject is uploaded to a server or cloud base storage. In some embodiments, the removing of the ear wax is conducted before providing the one or more audio stimuli to the subject’s ear. In some embodiments, the one or more audio stimuli comprise pure tones. In some embodiments, the one or more responses of the subject comprise pressing a surface when the subject detects or hears the one or more audio stimuli. In some embodiments, the surface comprises a surface of an interface on a computing device. In some embodiments, the computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, the recommendation comprises a referral to a health care professional. In some embodiments, the recommendation is provided to the subject on the computing device, a different personal device, website portal, personal computer, or any combination thereof. In some embodiments, the one or more audio stimuli are provided to the subject’s ear through a headphone or ear bud. In some embodiments, the headphone or the ear bud are wired or wireless. In some embodiments, determining the hearing classification of the subject comprises providing the one or more responses of the subject as an input to one or more trained machine learning algorithms or one or more predictive models that output the hearing classification of the subject. In some embodiments, the one or more trained machine learning algorithms or the one or more predictive models are trained with one or more responses of a plurality of subjects in response to one or more audio stimuli and a corresponding hearing classification of the plurality of subjects. In some embodiments, the one or more audio stimuli comprise audio stimuli of one or more frequencies and / or one or more intensities. Each audio stimulus may have an associated frequency and intensity. In some embodiments, an audio stimulus of the one or more audio stimuli is provided to the subject at a frequency of the one or more frequencies with at least 20 occurrences, and wherein each occurrence of the at least 20 occurrences comprises a different intensity of the audio stimulus. In some embodiments, the one or more frequencies comprise up to about 4 frequencies (e.g. 1, 2, 3 or 4 frequencies) or up to about 8 frequencies (e.g. 5, 6, 7, or 8 frequencies). In some embodiments, the one or more frequencies comprise: about 125Hz, about 250Hz, about 400Hz, about 750Hz, about 1000Hz, about 1500Hz, about 2000Hz, about 4000Hz, about 6000Hz, about 8000Hz, about 10,000Hz, about 12,500Hz, about 16,000Hz, one or more frequencies in the range 125 Hz to 16,000 Hz, or any combination thereof. In some embodiments, the one or more frequencies comprise one or more frequency bands. In some embodiments, the one or more frequency bands comprise frequency ranges of about 100Hz to about 999Hz, about 1000Hz to about 3999Hz, about 4000Hz to about 8000Hz, or any combination thereof. The one or more frequency bands may comprise frequencies of at least around 100 Hz, at least around 1000 Hz or at least around 4000 Hz, and / or the one or more frequency bands may comprise frequencies of no greater than around 999 Hz, no greater than around 3999 Hz or no greater than around 8000 Hz. In some embodiments, the subject’s hearing classification is determined from the subject’s one or more responses to at least 4 frequencies of the one or more frequencies of the one or more audio stimuli. In some embodiments, the subject’s hearing classification is determined from an average of the subject’s one or more responses from at about least 2 frequency bands, at least about 3 frequency bands, or at least 4 frequency bands of the one or more frequencies of the one or more audio stimuli. In some embodiments, the one or more responses from the subject comprise an intensity of the one or more audio stimuli at which the subject provided a response of detecting the one or more audio stimuli. In some embodiments, the subject’s hearing classification comprises normal hearing when the intensity comprises up to about 20dB, up to about 25dB, or up to about 40dB. In some embodiments, the hearing classification of the subject comprises normal hearing when the intensity of the one or more audio stimuli comprises about -lOdB to about 25dB. In some embodiments, the hearing classification of the subject comprises mild hearing loss when the intensity of the one or more audio stimuli comprises about 26dB to about 39dB. In some embodiments, the hearing classification of the subject comprises moderate hearing loss when the intensity of the one or more audio stimuli comprises about 40dB to about 69dB. In some embodiments, the hearing classification of the subject comprises severe hearing loss when the intensity of the one or more audio stimuli comprises about 70dB to about 94dB. In some embodiments, the hearing classification of the subject comprises profound hearing loss when the intensity of the one or more audio stimuli comprises about 95dB to about 120dB, or at least around 95 dB. The hearing classification of the subject may comprise hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of at least around 26 dB, at least around 40 dB, at least around 70 dB and / or at least around 95 dB. In some embodiments, determining the hearing classification comprises determining whether the subject has one or more of: normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, and / or an unfamiliar hearing profile. In some embodiments, the subject’s ear comprises a first ear and a second ear, wherein the one or more responses comprise a first set of one or more responses from the one or more audio stimuli provided to the first ear and a second set of one or more responses from providing the one or more audio stimuli to the second ear. In some embodiments, the subject’s hearing classification comprises asymmetric hearing when an intensity of at least two responses of the one or more responses of the subject’s first ear differ by at least about 15dB, or at least about 20dB from an intensity of at least two responses from the one or more responses of the subject’s second ear. For example, the subject’s hearing classification may comprise asymmetric hearing when, for each of two or more frequencies, intensity thresholds of the responses at the respective frequency differ by at least 15 dB between the first and second ears of the subject. In other words, the hearing classification may comprise asymmetric hearing when the first set of responses for the subject’s first ear comprises a first intensity threshold (e.g. minimum detected intensity of audio stimuli) at a first frequency and a second intensity threshold at a second frequency, and the second set of responses for the subject’s second ear comprises a first intensity threshold at the first frequency and a second intensity threshold at the second frequency, wherein the first intensity thresholds of the first and second ears at the first frequency differ by at least about 15 dB or at least about 20 dB, and the second intensity thresholds of the first and second ears at the second frequency differ by at least about 15 dB or at least about 20 dB. In some embodiments, determining the hearing classification of the subject comprises comparing the one or more responses of the subject to a library of one or more responses to the one or more audio stimuli associated with one or more hearing classifications. In some embodiments, the period of time comprises at least about 2 seconds or at least about 3 seconds. In some embodiments, the recommendation of the subject is uploaded to a server or cloud base storage. In some embodiments, the removing of the ear wax is conducted before providing the one or more audio stimuli to the subject’s ear. In some embodiments, the one or more audio stimuli comprise pure tones. In some embodiments, the one or more responses of the subject comprise pressing a surface when the subject detects or hears the one or more audio stimuli. In some embodiments, the surface comprises a surface of a user interface on the computing device. In some embodiments, the computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, the one or more programs further comprises instructions to generate or update an auditory health profde of the subject based on one or more of the determined hearing classification or the auditory health recommendation provided based thereon. In some embodiments, the subject’s hearing classification is determined by: providing one or more audio stimuli to the subject’s ear with a computing device; receiving one or more responses from the subject in response to the one or more audio stimuli with the computing device; and determining the hearing classification of the subject from the one or more responses. In some embodiments, the computing system comprises a smart phone, a server, a web portal, a personal computer, a laptop computer, a tablet, or any combination thereof. In some embodiments, the computing system is coupled to a headphone or ear bud. In some embodiments, the one or more responses from the subject comprise an intensity of the one or more audio stimuli at which the subject provided a response of detecting or hearing the one or more audio stimuli. In some embodiments, the determining the hearing classification comprises determining whether the subject has normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, or an unfamiliar hearing profile. In some embodiments, the subject’s hearing classification is further determined by removing the subject’s ear wax before providing the one or more audio stimuli to the subject’s ear. In some embodiments, the one or more responses of the subject comprise pressing a surface when the subject detects or hears the one or more audio stimuli. In some embodiments, the surface comprises a surface of a user interface on the computing device. In some embodiments, the computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, the determining the hearing classification of the subject comprises providing the one or more responses of the subject as an input to one or more trained machine learning algorithms or one or more predictive models that output the hearing classification of the subject. In some embodiments, the hearing referral comprises clinical data of the subject, contact information of the subject, the hearing classification, or any combination thereof. In some embodiments, the subject’s hearing classification is determined by: providing one or more audio stimuli to the subject’s ear with a computing device; receiving one or more responses from the subject in response to the one or more audio stimuli with the computing device; and determining the hearing classification of the subject from the one or more responses. In some embodiments, the computing system comprises a smart phone, a server, a web portal, a personal computer, a laptop computer, a tablet, or any combination thereof. In some embodiments, the computing system is coupled to a headphone or ear bud. In some embodiments, the one or more responses from the subject comprise an intensity of the one or more audio stimuli at which the subject provided a response of detecting or hearing the one or more audio stimuli. In some embodiments, determining the hearing classification comprises determining whether the subject has normal hearing, normal hearing, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, or an unfamiliar hearing profile. In some embodiments, the subject’s hearing classification is further determined by removing the subject’s ear wax before providing the one or more audio stimuli to the subject’s ear. In some embodiments, the one or more responses of the subject comprise pressing a surface when the subject detects or hears the one or more audio stimuli. In some embodiments, the surface comprises a surface of a user interface on the computing device. In some embodiments, the computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, determining the hearing classification of the subject comprises providing the one or more responses of the subject as an input to one or more trained machine learning algorithms or one or more predictive models that output the hearing classification of the subject. In some embodiments, the hearing referral comprises clinical data of the subject, contact information of the subject, the hearing classification, or any combination thereof. In some embodiments, the instructions further comprise accepting the subject’s hearing referral. The present disclosure provides a method for training a machine learning algorithm or predictive model, comprising: receiving or obtaining one or more training subject responses to one or more audio stimuli provided to the one or more training subjects’ ear(s) and one or more corresponding hearing classifications of the one or more training subjects; and training the machine learning algorithm or predictive model with the one or more responses and corresponding hearing classifications thereby generating a trained machine learning algorithm or trained predictive model. In some embodiments, the one or more training subject responses to the one or more audio stimuli comprise an intensity of the one or more audio stimuli that the one or more training subjects detect or hear the one or more audio stimuli. In some embodiments, the machine learning algorithm or predictive model comprises a neural network, a support vector machine, a random forest model, a naive Bayes classification algorithm, a gradient-boosting algorithm, linear regression algorithm, unsupervised clustering algorithm, supervised clustering algorithm, or any combination thereof. In some embodiments, the trained machine learning algorithm or the trained predictive model is configured to predict or determine one or more subjects’ hearing classification with at least about 85% accuracy, specificity, sensitivity, or any combination thereof. In some embodiments, the one or more training subject responses comprise one or more responses of the one or more training subjects to the one or more audio stimuli provided to a left ear and a right ear of the one or more training subjects’ ears. In some embodiments, the one or more audio stimuli are provided to the one or more training subjects by a computing device. In some embodiments, the computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, the one or more training subject responses comprise pressing a surface when the one or more training subjects detect or hear the one or more audio stimuli. In some embodiments, the surface comprises a surface of the computing device. In some embodiments, the trained machine learning algorithm or the trained predictive model is configured to receive an input of a subject’s one or more responses of detecting or hearing one or more audio stimuli provided to the subject’s ear, and wherein the trained machine learning algorithm or the trained predictive model is configured to output a hearing classification of the subject. In some embodiments, the trained machine learning algorithm or the trained predictive model comprises a plurality of trained machine learning algorithms or a plurality of trained predictive models.

[0004] In general, the decibel (“dB”) values of detected intensity levels provided herein are defined in “decibels in hearing level” (dB HL), which expresses the intensity of an audio stimulus relative to an average reference threshold of hearing in normal-hearing young adults. Accordingly, the intensity of the audio stimuli described herein may be presented in dB HL, such that hearing classifications reflect subjects’ auditory sensitivity relative to normative hearing thresholds. However, in some implementations, the “dB” levels provided herein may be defined in another decibel measure of intensity, such as “decibel sound pressure level” (dB SPL), which is a physical measure of sound intensity based on local pressure deviations from atmospheric pressure caused by the sound wave. Intensities measured in dB SPL may refer to the raw absolute output produced by the sound source (e.g. headphone), for example based on its engineering specifications. For example, the physical outputs of audio devices may be measured and controlled in dB SPL, which may be converted to perceived intensity levels in dB HL based on calibration data (e.g. reference equivalent sound pressure levels, RETSPLs) for transducers of the audio device. Other measures of sound intensity include decibels relative to full scale (dB FS) which defines sound intensity levels relative to a maximum digital level. Digital audio stimuli (e.g. sine waves) may be defined and generated in dB FS, which may be converted to a dB SPL measure by applying an appropriate correction factor based on the frequency and transducer specifications.

[0005] The present disclosure provides a system for providing an auditory health recommendation to a subject, comprising: one or more processors and memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions to perform the method of any one of claims 103-113.

[0006] The present disclosure provides a system configured to train a machine learning algorithm or predictive model, comprising: one or more processors and memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions to: receive or obtain one or more training subject responses to one or more audio stimuli provided to the one or more training subjects’ ear(s) and corresponding one or more hearing classification of the one or more training subjects; and train the machine learning algorithm or predictive model with the one or more responses and corresponding hearing classifications thereby generating a trained machine learning algorithm or trained predictive model. In some embodiments, the one or more training subject responses to one or more audio stimuli comprise an intensity of the one or more audio stimuli that the one or more training subjects detect or hear the one or more audio stimuli. In some embodiments, the machine learning algorithm or predictive model comprises a neural network, support vector machine, random forest, naive Bayes classification algorithm, gradient-boosting algorithm, linear regression algorithm, unsupervised clustering algorithm, supervised clustering algorithm, or any combination thereof. In some embodiments, the trained machine learning algorithm or the trained predictive model is configured to predict or determine one or more subjects’ hearing classification with at least about 85% accuracy, specificity, sensitivity, or any combination thereof. In some embodiments, the one or more training subject responses comprise one or more responses of the one or more training subjects to the one or more audio stimuli provided to a left ear and a right ear of the one or more training subjects’ ears. In some embodiments, the one or more audio stimuli are provided to the one or more training subjects by a computing device. In some embodiments, the computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, the one or more training subject responses comprise pressing a surface when the one or more training subjects detect or hear the one or more audio stimuli. In some embodiments, the surface comprises a surface of the computing device. In some embodiments, the trained machine learning algorithm or the trained predictive model is configured to receive an input of a subject’s one or more response of detecting or hearing one or more audio stimuli provided to the subject, and wherein the trained machine learning algorithm or the trained predictive model is configured to output a hearing classification of the subject. In some embodiments, the trained machine learning algorithm or the trained predictive model comprises a plurality of trained machine learning algorithms or a plurality of trained predictive models.

[0007] In some embodiments, the present disclosure provides systems and / or methods that may evaluate a subject’s hearing through the use of a computing device (e.g., a smartphone), corresponding hardware (e.g., headphones and / or ear buds) to deliver one or more auditory stimuli, and / or an application on the computing device that provides the one or more auditory stimuli and / or records, receives, and / or detects a subject’s response to the one or more auditory stimuli. In some embodiments, the application on the computing device may comprise e.g., a mobile application that may be used by clinicians in point-of-care and / or point-of-need settings such as pharmacies, doctor’s offices, and / or other community health settings.

[0008] Another aspect of the present disclosure provides a system comprising one or more computer processors and computer memory coupled thereto. The computer memory may comprise machine executable code that, upon execution by the one or more computer processors, implements any of the methods described elsewhere herein.

[0009] Additional aspects and advantages of the present disclosure will become readily apparent to those skilled in this art from the following detailed description, where only illustrative embodiments of the present disclosure are shown and described. As will be realized, the present disclosure is capable of other and different embodiments, and its several details are capable of modifications in various obvious respects, all without departing from the disclosure. Accordingly, the drawings and description are to be regarded as illustrative in nature, and not as restrictive. INCORPORATION BY REFERENCE

[0010] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference. To the extent publications and patents or patent applications incorporated by reference contradict the disclosure contained in the specification, the specification is intended to supersede and / or take precedence over any such contradictory material. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present disclosure will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the present disclosure are utilized, and the accompanying drawings (also “Figure” and “FIG.” herein), of which:

[0012] FIG. 1 shows a flow diagram of the various user interface views of the computing device application when the computing device is performing a hearing test, as described in some embodiments herein.

[0013] FIGS. 2A-2C show user interface views of the computing device application after a hearing test is conducted, as described in some embodiments herein. FIG. 2A shows an assessment of both ear audiogram values with academic sources. FIG. 2B shows common audiogram trend information. FIG. 2C shows an overlay of common audiogram trends onto the subject’s audiogram.

[0014] FIGS. 3A-3C show user interface views of the computing device application after a hearing test and data analysis are performed to assess a subject’s auditory health, as described in some embodiments herein. FIG. 3A shows a subject’s audiogram that is found to be within normal limits. FIG. 3B shows a subject’s audiogram where the right ear has been found to be outside normal limits, and the left ear has been found to be normal. FIG. 3C shows a subject’s audiogram where the right ear has been found to be inconclusive and the left ear has been found to be normal.

[0015] FIG. 4A shows a user interface view of the computing device application displaying a collapsible menu of common audiogram trends that can be scrolled through with user interface objects that route a subject to a different user interface view with additional information, as described in some embodiments herein.

[0016] FIG. 4B shows a user interface view of the computing device application with additional information pertaining to the common audiogram trend including, e.g., a description of the trend, clinical advice, subject advice, and / or references, as described in some embodiments herein.

[0017] FIG. 5A shows an interactive user interface view of the computing device application where users can click, touch, and / or tap the tiles of the common audiogram trends to produce an overlay on the subject’s audiogram, as described in some embodiments herein.

[0018] FIG. 5B shows a user interface view of the computing device application displaying questions that a user is asked when making a final judgment of the subject’s hearing loss for each ear, as described in some embodiments herein.

[0019] FIG. 6A shows a user interface view of the computing device application displayed when a user selects both ears as “within normal limits” or “N / A”, as described in some embodiments herein.

[0020] FIG. 6B shows a user interface view of the computing device application when a user has selected either ear as “recommended for detailed hearing assessment”, as described in some embodiments herein.

[0021] FIGS. 7A-7C shows user interface views of the computing device application pertaining to completing an auditory health referral, as described in some embodiments herein. FIG. 7A shows a user interface view of the computing device application displaying instructions for the referral and input boxes where the subject can input their information. FIG. 7B shows a user interface view of the computing device application displaying a list of organizations presented to the subject for an auditory health referral. FIG. 7C shows data and / or referral sync user interface view of the computing device application when the computing device is not connected to the internet.

[0022] FIG. 8 shows a flow diagram of user interface views of the computing device application for providing an auditory health referral, as described in some embodiments herein. Steps for providing an auditory health referral are shown sequentially.

[0023] FIGS. 9A-9B shows a user interface view of a cloud-based computing system displaying auditory health referrals, related subject information, and / or referral appointment follow up information, as described in some embodiments herein.

[0024] FIG. 10 shows a flow diagram of the subject auditory health referral process from the perspective of a user and / or a clinician, as described in some embodiments herein.

[0025] FIG. 11 shows a flow diagram of the subject auditory health referral process from the point of view of the referral organization, as described in some embodiments herein.

[0026] FIG. 12 shows a graph of auditory stimulation intensity of a repeated tone and corresponding response from a subject, as described in some embodiments herein.

[0027] FIG. 13A shows a home screen user interface view of the computing device application, as described in some embodiments herein.

[0028] FIG. 13B shows a user interface view of the computing device application displaying a list of appointments for a user, as described in some embodiments herein.

[0029] FIG. 13C shows a user interface view of the computing device application displaying a calendar feature where a user can select a day on the calendar to see their appointments for that day, as described in some embodiments herein.

[0030] FIG. 13D shows a user interface view of the computing device application displaying a list of subjects, as described in some embodiments herein.

[0031] FIG. 13E shows a user interface view of the computing device application displaying a sync screen indicating whether data from each appointment has been synchronized with the internet, as described in some embodiments herein.

[0032] FIG. 13F shows a user interface view of the computing device application displaying a home screen of the computing device application and an image of the otoscopy device, as described in some embodiments herein.

[0033] FIGS. 14A-14D show user interface views of the computing device application displaying a subject’s audiogram and artificial intelligence (AI) analysis and / or processing of the subject’s audiogram, as described in some embodiments herein.

[0034] FIG. 15 shows a computer system that is programmed or otherwise configured to implement methods described elsewhere herein, as described in some embodiments herein. DETAILED DESCRIPTION

[0035] Provided herein is a method of providing an auditory health recommendation to a subject, comprising: (i) providing one or more audio stimuli to a subject’s ear with a computing device; (ii) receiving one or more responses from the subject in response to the one or more audio stimuli with the computing device; (iii) determining a hearing classification of the subject from the one or more responses; and (iv) providing the auditory health recommendation to the subject based on the hearing classification. In some embodiments, steps (i) to (iv) may be repeated at one or more regular intervals. In some examples, step (i) may be optional.

[0036] The computing device application can provide a rules-based approach for determining if hearing thresholds of a subject are within normal limits and / or outside of normal limits. For example, a classification of normal hearing may indicate that hearing thresholds of the subject are within normal limits for a range of frequencies tested using the one or more audio stimuli, although this may not exclude the possibility of hearing loss at non-tested frequencies, meaning a hearing classification indicative of hearing loss may be determined for the same subject (e.g. subsequently or concurrently with the normal hearing classification) for a different range of frequencies. Accordingly, the hearing classification may be associated with a range of frequencies of the one or more audio stimuli.

[0037] In some instances, a normal hearing classification threshold of a subject may be less than or equal to about 25 dB intensity of one or more audio stimuli provided at one or more frequencies and / or one or more frequency bands of the one or more audio stimuli to a left and / or a right ear of a subject where a subject provides a response of detecting and / or hearing the one or more audio stimuli. Hearing classifications outside of normal limits may be determined as greater than about 25 dB intensity of one or more audio stimuli provided at one or more frequencies and / or one or more frequency bands of the one or more audio stimuli to a left and / or right ear of a subject where a subject provides a response of detecting and / or hearing the one or more audio stimuli. In some cases, a subject may be classified with a hearing classification of normal when the subject provides a response to detecting and / or hearing one or more audio stimuli at about 15 dB to about 25 dB intensity of the one or more auditory stimuli across one or more frequencies and / or one or more frequency bands of the one or more audio stimuli. In some instances, a subject may be classified with a hearing classification of normal hearing when the subject provides a response to detecting and / or hearing the one or more audio stimuli at up to about 25 dB intensity of the one or more auditory stimuli across one or more frequencies and / or one or more frequency bands of the one or more audio stimuli. The hearing test administered and / or provided by the methods and / or systems, described elsewhere herein, may be performed on a subject in non-soundproof environment. In some instances, to account for potential errors (e.g., arising from ambient noise in an environment) in assessing and / or determining a subject’s hearing classification and / or administering and / or providing a hearing test, as described elsewhere herein, a threshold intensity of about 20 dB to about 60 dB of one or more audio stimuli across one or more frequencies and / or one or more frequency bands of the one or more audio stimuli that a subject provides a response of detecting and / or hearing the one or more audio stimuli may be utilized. In some cases, a subject may be determined to be a candidate for hearing aids if the subject’s hearing threshold of the one or more audio stimuli is up to about 25dB.

[0038] In some cases, a hearing classification of hearing asymmetry may comprise about a 15 dB intensity difference at one or more frequencies and / or one or more frequency bands of one or more audio stimuli provided to a subject where the subject provides a response of detecting and / or hearing the one or more audio stimuli. In some cases, a hearing classification of hearing asymmetry may be determined by providing a subject up to about four frequencies of the one or more audio stimuli, described elsewhere herein. In some cases, a hearing classification of hearing asymmetry may be determined by a difference of about 15 dB intensity of one or more audio stimuli provided to a subject where the subject provides a response of detecting and / or hearing the one or more audio stimuli, where the one or more audio stimuli are provided at about two adjacent frequencies and / or frequency bands. In some cases, the about two adjacent frequencies may comprise a frequency of about 0.5 kHz, about 1 kHz, about 2 kHz, about 4 kHz, and / or about 8 kHz. In some cases, a hearing classification of a subject may comprise asymmetry hearing when the subject provides a response of detecting and / or hearing an intensity of one or more audio stimuli at greater than or equal to about 20 dB at two contiguous frequencies of the one or more audio stimuli or greater than or equal to about 10 dB at three contiguous frequencies of the one or more frequencies and / or one or more frequency bands, described elsewhere herein, of the one or more audio stimuli between a left and a right ear of the subject.

[0039] In some cases, the one or more audio stimuli at the one or more frequencies and / or one or more frequency bands may be provided to a subject at one or more intensities (e.g., volume) to obtain a subject’s response to hearing and / or detecting the one or more audio stimuli, as shown in FIG. 12. For example, FIG. 12 shows a graph of a sequence of one or more audio stimuli at a frequency of the one or more frequencies provided to a subject where a portion of the one or more audio stimuli at the varying intensity are detected and / or heard by the subject where others are not. In some cases, a frequency of the one or more audio stimuli may be provided to a subject up to about 20 instances per frequency to determine a subject’s response to the audio stimuli at the frequency. In some cases, if no response (e.g., a threshold hearing intensity at the frequency) from the subject is established by the 20th instance, the hearing test, as described elsewhere herein, may proceed with providing an audio stimulus of the next frequency. In some cases, an inconclusive response (e.g., indicated by a question mark object) may be displayed on the subject’s hearing threshold audiogram, described elsewhere herein, where the subject did not provide a sufficient number of responses to the audio stimuli.

[0040] In some cases, the methods and / or systems, described elsewhere herein, may support a user in making informed decisions about a subject’s auditory health by providing clear, accessible, and / or relevant information of a subject’s hearing test results. Users may be clinicians, including both specialists and non-specialists, who have completed minimum training requirements from an academy for training clinicians on how to perform hearing tests.

[0041] In some embodiments, the methods described herein may be used in community health settings, such as pharmacies, outpatient clinics, family medicine outpatient clinics, urgent care, pharmacy clinics, or any combination thereof. In some embodiments, users can compare a subject’s hearing results against reference hearing test results to determine hearing loss. In some cases, the methods and / or systems, described elsewhere herein may utilizing artificial intelligence (e.g., one or more machine learning algorithms and / or one or more predictive models) to assess a subject’s hearing classification from the subject’s audiogram. In some instances, the methods and / or systems, described elsewhere herein, may be utilized as an adjunctive and / or supplemental method and / or system to other clinical diagnostic tools, methods, and / or prognostic indicators of determining a subject’s hearing classification and / or auditory health. In some cases, a hearing classification output and / or determination of the methods and / or systems, described elsewhere herein, may provide educational information and / or supplemental information related to a subject’s hearing classification that assists a knowledge base of a user.

[0042] In some embodiments, the auditory health recommendation provided to a subject may comprise a referral to a health care professional. In some embodiments, the auditory health recommendation may be provided to the subject on the computing device, a different personal computing device, website portal, personal computer, or any combination thereof. In some embodiments, one or more audio stimuli may be provided to the subject’s ear through a headphone or ear bud. In some embodiments, the headphone or the ear bud may be wired or wireless. In some embodiments, determining hearing classification of the subject may comprise providing the one or more responses of the subject to hearing and / or detecting the one or more audio stimuli as an input to one or more trained machine learning algorithms and / or one or more predictive models, where the one or more trained machine learning algorithms and / or the one or more predictive models output the hearing classification of the subject. The method may use an algorithmic approach to determine whether a result per ear is outside normal hearing limits or within normal hearing limits, as described elsewhere herein. When there is an inconclusive or no response, the one or more machine learning algorithms and / or the one or more predictive models may provide, display, indicate, and / or issue a flag per ear for review. The displayed result per ear may be interpreted by the clinician in the clinical context.

[0043] In some cases, the user may use the methods and / or systems, described elsewhere herein, to compare a subject’s hearing results, e.g., a subject’s hearing classification, against reference hearing test result and / or hearing classifications to determine hearing loss and underlying pathology. The method may be used to inform and empower a non-specialist user about different audiogram types, described elsewhere herein. The method may be used to compare a subject’s hearing test result and / or hearing classification with standard reference audiograms.

[0044] In some embodiments, a user may be notified using this method that the results from the hearing test (e.g. hearing thresholds or detected intensities of audio stimuli) are within normal or outside of normal limits. The methods and systems, described elsewhere herein, may provide a user a notification of the hearing classification of a left and right ear of the subject. A user may be assisted in understanding the potential ear-specific underlying issues through information and / or educational literature provided to the user in connection with the subject’s hearing classification and / or hearing test results. A user may be able to find out more about a hearing result profile through interacting with one or more user interface views of an application of a computing device.

[0045] One or more user interface views of the computing device application may comprise displaying and / or indicating a hearing classification and / or notification with one or more colored objects and / or color objects overlaid over one or more other user interface view objects and / or text e.g., to indicate when a subject’s hearing classification and / or hearing test results are outside of normal hearing limits. In some instances, one or more user interface views of the computing device application may comprise black and white (e.g., varying levels of grayscale and / or dotted line type) objects to, e.g., indicate a subject’s hearing classification and / or hearing test results. In some cases, one or more user interface views of the computing device may display one or more objects to indicate when artificial intelligence, e.g., one or more machine learning models and / or one or more predictive models are in use and / or have been used to generate and / or process a subject’s audiogram and / or responses. In some cases, the one or more objects to indicate when artificial intelligence is used may comprise one or more symbols.

[0046] In some embodiments, the one or more trained machine learning algorithms and / or the trained one or more predictive models may be trained with one or more responses of a plurality of subjects in response to one or more audio stimuli and a corresponding hearing classification of the plurality of subjects. In some embodiments, the one or more audio stimuli may comprise audio stimuli of one or more frequencies and / or one or more intensities. Each audio stimulus may have an associated frequency and intensity. In some embodiments, an audio stimulus of the one or more audio stimuli may be provided to the subject at a frequency of the one or more frequencies with at least 20 occurrences, and each occurrence of the at least 20 occurrences may comprise a different intensity of the audio stimulus. In some embodiments, the one or more frequencies may comprise up to about 4 frequencies (e.g. 1, 2, 3 or 4 frequencies) or up to about 8 frequencies (e.g. 5, 6, 7, or 8 frequencies). In some embodiments, the one or more frequencies may comprise: about 250 Hz, about 500 Hz, about 1000 Hz, about 2000 Hz, about 3000 Hz, about 4000 Hz, about 6000 Hz, about 8000 Hz, one or more frequencies in the range 250 Hz to 8000 Hz, or any combination thereof. In some embodiments, the one or more frequencies may comprise one or more frequency bands. In some embodiments, the one or more frequency bands may comprise frequency ranges of about 100 Hz to about 999 Hz, about 1000 Hz to about 3999 Hz, about 4000 Hz to about 8000 Hz, or any combination thereof. The one or more frequency bands may comprise frequencies of at least around 100 Hz, at least around 1000 Hz or at least around 4000 Hz, and / or the one or more frequency bands may comprise frequencies of no greater than around 999 Hz, no greater than around 3999 Hz or no greater than around 8000 Hz. In some embodiments, the subject’s hearing classification may be determined from the subject’s one or more responses to at least 4 frequencies of the one or more frequencies of the one or more audio stimuli. In some embodiments, the subject’s hearing classification may be determined from an average of the subject’s one or more responses from at least about 2 frequencies and / or at least about 2 frequency bands, at least about 3 frequencies and / or at least about 3 frequency bands, or at least 4 frequencies and / or at least 4 frequency bands of the one or more frequencies of the one or more audio stimuli. In some embodiments, the one or more responses from the subject may comprise an intensity of the one or more audio stimuli at which the subject provided a response of detecting and / or hearing the one or more audio stimuli. In some embodiments, the subject’s hearing classification may comprise normal hearing when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of up to about 20 dB, up to about 25 dB, or up to about 40 dB. In some embodiments, the hearing classification of the subject may comprise normal hearing when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of about -10 dB to about 25 dB. In some embodiments, the hearing classification of the subject may comprise mild hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity an intensity of about 26 dB to about 39 dB. In some embodiments, the hearing classification of the subject may comprise moderate hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of about 40 dB to about 69 dB. In some embodiments, the hearing classification of the subject may comprise severe hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of about 70 dB to about 94 dB. In some embodiments, the hearing classification of the subject may comprise profound hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises about 95 dB to about 120 dB. or at least around 95 dB. The hearing classification of the subject may comprise hearing loss (or abnormal hearing) when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of at least around 26 dB, at least around 40 dB, at least around 70 dB and / or at least around 95 dB. In some embodiments, determining the hearing classification may comprises determining whether the subject has normal hearing, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, or an unfamiliar hearing profile. In some embodiments, the subject’s ears may comprise a first ear and a second ear, and the one or more responses may comprise a first set of one or more responses from the one or more audio stimuli provided to the first ear and a second set of one or more responses from providing the one or more audio stimuli to the second ear. In some embodiments, the subject’s hearing classification may comprise asymmetric hearing when an intensity of the one or more audio stimuli of at least about two responses of the one or more responses of the subject’s first ear differ by at least about 15 dB, or at least about 20 dB from an intensity of the one or more audio stimuli of at least two responses from the one or more responses of the subject’s second ear. For example, the subject’s hearing classification may comprise asymmetric hearing when, for each of two or more frequencies, intensity thresholds of the responses at the respective frequency differ by at least 15 dB between the first and second ears of the subject. In other words, the hearing classification may comprise asymmetric hearing when the first set of responses for the subject’s first ear comprises a first intensity threshold (e.g. minimum detected intensity of audio stimuli) at a first frequency and a second intensity threshold at a second frequency, and the second set of responses for the subject’s second ear comprises a first intensity threshold at the first frequency and a second intensity threshold at the second frequency, wherein the first intensity thresholds of the first and second ears at the first frequency differ by at least about 15 dB or at least about 20 dB, and the second intensity thresholds of the first and second ears at the second frequency differ by at least about 15 dB or at least about 20 dB.In some embodiments, the method may further comprise displaying an object overlaid on a graph of the one or more responses of the subject when a reference audiogram is selected. In some embodiments, determining the hearing classification of the subject may comprise comparing the one or more responses of the subject to a library of one or more responses to the one or more audio stimuli associated with one or more hearing classifications. In some embodiments, the method of determining a subject’s hearing classification and / or conducting a hearing test on a subject may further comprise providing a period of time where no audio stimulation is provided to the subject’s ear. In some embodiments, the period of time may comprise at least about 2 seconds or at least about 3 seconds. In some embodiments, the auditory health recommendation of the subject may be uploaded to a server or cloud base storage. In some embodiments, the method may further comprise removing ear wax from the subject’s ear. In some embodiments, the removing of the ear wax may be conducted before and / or after providing the one or more audio stimuli to the subject’s ear. In some cases, the method may further comprise removing ear wax from a subject’s ear at least as a result of determining and / or obtaining an inconclusive hearing classification of a subject for a left ear and / or a right ear of the subject. In some embodiments, the one or more audio stimuli may comprise pure tones (e.g. sine waves). In some embodiments, the one or more responses of the subject to the one or more audio stimuli may comprise pressing a surface when the subject detects or hears the one or more stimuli. In some embodiments, the surface may comprise a surface of an interface (e.g., a user interface view) on a computing device. In some embodiments, the computing device may comprise a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, the method may further comprise generating or updating an auditory health profde of the subject based on one or more of the determined hearing classifications or auditory health recommendations provided based thereon.

[0047] Provided herein, in some embodiments, is a system for providing an auditory health recommendation to a subject, comprising: one or more processors and memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions to: (i) provide one or more audio stimuli to the subject’s ear with a computing device; (ii) receive one or more responses from the subject in response to the one or more audio stimuli with the computing device; (iii) determine a hearing classification of the subject from the one or more responses with the one or more processors; and (iv) provide the auditory health recommendation to the subject based on the hearing classification. In some embodiments, the system program instructions may further comprise repeating steps (i) to (iv) at one or more regular intervals. In some embodiments, the recommendation may comprise a referral to a health care professional. In some embodiments, the recommendation may be provided to the subject on the computing device, a different personal device, website portal, personal computer, or any combination thereof. In some embodiments, the one or more audio stimuli may be provided to the subject’s ear through a headphone or ear bud. In some embodiments, the headphone or the ear bud may be wired or wireless. In some embodiments, determining the hearing classification of the subject may comprise providing the one or more responses of the subject as an input to one or more trained machine learning algorithms and / or one or more predictive models that output the hearing classification of the subject. In some embodiments, the one or more trained machine learning models or the trained one or more predictive models may be trained with one or more responses of a plurality of subjects in response to one or more audio stimuli and a corresponding hearing classification of the plurality of subjects. In some embodiments, the one or more audio stimuli may comprise audio stimuli of one or more frequencies and / or one or more intensities. Each audio stimulus may have an associated frequency and intensity. In some embodiments, an audio stimulus of the one or more audio stimuli may be provided to the subject at a frequency of the one or more frequencies with at least 20 occurrences, and where each occurrence of the at least 20 occurrences may comprise a different intensity of the audio stimulus. In some embodiments, the one or more frequencies may comprise up to about 4 frequencies (e.g. 1, 2, 3 or 4 frequencies) or up to about 8 frequencies (e.g. 5, 6. 7, or 8 frequencies). In some embodiments, the one or more frequencies may comprise: about 125Hz, about 250Hz, about 400Hz, about 750Hz, about 1000Hz, about 1500Hz, about 2000Hz, about 4000Hz, about 6000Hz, about 8000Hz, about 10,000Hz, about 12,500Hz, about 16,000Hz, one or more frequencies in the range 125 Hz to 16,000 Hz, or any combination thereof. In some embodiments, the one or more frequencies may comprise one or more frequency bands. In some embodiments, the subject’s hearing classification may be determined from the subject’s one or more responses to at least about 4 frequencies and / or at least about 8 frequencies of the one or more frequencies of the one or more audio stimuli. In some instances, the at least about 4 frequencies may comprise about 500 Hz, about 1000 Hz, about 2000 Hz, and about 4000 Hz, and / or the at least about 4 frequencies may be at least around 500 Hz and / or no greater than around 4000 Hz. In some cases, the at least about 8 frequencies may comprise a frequency of about 250 Hz, about 500 Hz, about 1000 Hz, about 2000 Hz, about 3000 Hz, about 4000 Hz, about 6000 Hz, and about 8000 Hz, and / or the at least about 8 frequencies may be at least around 250 Hz and / or no greater than around 8000 Hz. In some embodiments, the subject’s hearing classification may be determined from an average of the subject’s one or more responses from at least about 2 frequency bands, at least about 3 frequency bands, or at least 4 frequency bands of the one or more frequencies of the one or more audio stimuli In some embodiments, the one or more responses from the subject may comprise an intensity of the one or more audio stimuli at which the subject provided a response of detecting, identifying, and / or hearing the one or more audio stimuli, determined in absolute or relative terms. In some embodiments, the subject’s hearing classification may comprise normal hearing when the intensity of the one or more audio stimuli detected and / or identified by the subject (also referred to as one or more hearing thresholds) comprises an intensity of up to about 20 dB, up to about 25 dB, or up to about 40 dB. In some embodiments, the hearing classification of the subject may comprise normal hearing when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of about -10 dB to about 25 dB. In some embodiments, the hearing classification of the subject may comprise mild hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of about 26 dB to about 39 dB. In some embodiments, the hearing classification of the subject may comprise moderate hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of about 40 dB to about 69 dB. In some embodiments, the hearing classification of the subject may comprise severe hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of about 70 dB to about 94 dB. In some embodiments, the hearing classification of the subject may comprise profound hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of about 95 dB to about 120 dB, or at least around 95 dB. The hearing classification of the subject may comprise hearing loss when the intensity of the one or more audio stimuli detected and / or identified by the subject comprises an intensity of at least around 26 dB, at least around 40 dB, at least around 70 dB and / or at least around 95 dB. In some embodiments, determining the hearing classification of the subject may comprise determining whether the subject has normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, or an unfamiliar hearing profile. In some embodiments, the subject’s ears may comprise a first ear and a second ear, wherein the one or more responses may comprise a first set of one or more responses from the one or more audio stimuli provided to the first ear and a second set of one or more responses from providing the one or more audio stimuli to the second ear. In some embodiments, the subject’s hearing classification may comprise asymmetric hearing when an intensity of the one or more audio stimuli of at least two responses of the one or more responses of the subject’s first ear differ by at least about 15 dB, or at least about 20 dB from an intensity of the one or more audio stimuli of at least two responses from the one or more responses of the subject’s second ear. For example, the subject’s hearing classification may comprise asymmetric hearing when, for each of two or more frequencies, intensity thresholds of the responses at the respective frequency differ by at least 15 dB between the first and second ears of the subject. In other words, the hearing classification may comprise asymmetric hearing when the first set of responses for the subject’s first ear comprises a first intensity threshold (e.g. minimum detected intensity of audio stimuli) at a first frequency and a second intensity threshold at a second frequency, and the second set of responses for the subject’s second ear comprises a first intensity threshold at the first frequency and a second intensity threshold at the second frequency, wherein the first intensity thresholds of the first and second ears at the first frequency differ by at least about 15 dB or at least about 20 dB, and the second intensity thresholds of the first and second ears at the second frequency differ by at least about 15 dB or at least about 20 dB. In some embodiments, the system may further comprise displaying an object overlaid on a graph of the one or more responses of the subject when a reference audiogram is selected. In some embodiments, determining the hearing classification of the subject may comprise comparing the one or more responses of the subject to a library of one or more responses to the one or more audio stimuli associated with one or more hearing classifications. In some embodiments, the system may further comprise providing a period of time where no audio stimulation is provided to the subject’s ear(s). In some embodiments, the period of time may comprise at least about 2 seconds or at least about 3 seconds. In some embodiments, the recommendation of the subject may be uploaded to a server or cloud-based storage. In some embodiments, the system may further comprise removing ear wax from the subject’s ear. In some embodiments, the removing of the ear wax may be conducted before providing the one or more audio stimuli to the subject’s ear. In some embodiments, the one or more audio stimuli may comprise pure tones. In some embodiments, the one or more responses of the subject may comprise pressing a surface when the subject detects or hears the one or more audio stimuli. In some embodiments, the surface may comprise a surface of one or more user interface (e.g., of an application) on the computing device. In some embodiments, the computing device may comprise a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof.

[0048] Provided herein, in some embodiments, is a method for providing a subject’s hearing referral to a health care provider, comprising: providing the subject’s hearing referral to the health care provider from a computing system in response to the computer system receiving the subject’s hearing classification and indication for referral, where the subject’s hearing classification is determined by: providing one or more audio stimuli to the subject’s ear with a computing device; receiving one or more responses from the subject in response to the one or more audio stimuli with the computing device; and determining the hearing classification of the subject from the one or more responses. In some embodiments, the computing system may comprise a smart phone, a server, a web portal, a personal computer, a laptop computer, a tablet, or any combination thereof. In some embodiments, the one or more responses from the subject may comprise an intensity of the one or more audio stimuli at which the subject provided a response of detecting or hearing the one or more audio stimuli. In some embodiments, determining the hearing classification may comprise determining whether the subject has one or more of: normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, and / or an unfamiliar hearing profile. In some embodiments, the subject’s hearing classification may be further determined by removing the subject’s ear wax before and / or after providing the one or more audio stimuli to the subject’s ear. In some embodiments, the one or more responses of the subject may comprise pressing a surface when the subject detects, hears, and / or identifies the one or more audio stimuli. In some embodiments, the surface may comprise a surface of a user interface on the computing device. In some embodiments, the computing device may comprise a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, determining the hearing classification of the subject may comprise providing the one or more responses of the subject as an input to one or more trained machine learning algorithms and / or one or more predictive models that output the hearing classification of the subject. In some embodiments, the hearing referral may comprise clinical data of the subject, contact information of the subject, the hearing classification, or any combination thereof. In some embodiments, the method may further comprise accepting the subject’s hearing referral.

[0049] Provided herein, in some embodiments, is a system for providing a subject’s hearing referral to a health care provider, comprising: one or more processors and memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions to: provide the subject’s hearing referral to the health care provider from a computing system in response to the computing system receiving the subject’s hearing classification and indication for referral from a computing device, where the subject’s hearing classification is determined by: providing one or more audio stimuli to the subject’s ear with a computing device; receiving one or more responses from the subject in response to the one or more auditory stimuli with the computing device; and determining the hearing classification of the subject from the one or more responses. In some embodiments, the computing system may comprise a smart phone, a server, a web portal, a personal computer, a laptop computer, a tablet, or any combination thereof. In some embodiments, the one or more responses from the subject may comprise an intensity of the one or more audio stimuli at which the subject provided a response of detecting, hearing, and / or identifying the one or more audio stimuli. In some embodiments, determining the hearing classification may comprise determining whether the subject has one or more of: normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, and / or an unfamiliar hearing profile. In some embodiments, the subject’s hearing classification may be further determined by removing the subject’s ear wax before and / or after providing the one or more audio stimuli to the subject’s ear. In some embodiments, the one or more responses of the subject may comprise pressing a surface when the subject detects, hears, and / or identifies the one or more audio stimuli. In some embodiments, the surface may comprise a surface of a user interface (e.g., the user interface of an application) on the computing device. In some embodiments, the computing device may comprise a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, determining the hearing classification of the subject may comprise providing the one or more responses of the subject as an input to one or more trained machine learning algorithms and / or one or more predictive models that output the hearing classification of the subject. In some embodiments, the hearing referral may comprise clinical data of the subject, contact information of the subject, the hearing classification, or any combination thereof. In some embodiments, the instructions may further comprise accepting the subject’s hearing referral.

[0050] Provided herein, in some embodiments, is a method for training an untrained or partially untrained machine learning algorithm or predictive model, comprising: receiving or obtaining one or more training subject responses to one or more audio stimuli provided to the one or more training subjects’ ear(s) and one or more corresponding hearing classifications of the one or more training subjects; and training the untrained or the partially untrained machine learning algorithm and / or predictive model with the one or more responses and corresponding hearing classifications thereby generating a trained machine learning algorithm and / or trained predictive model. In some embodiments, the one or more training subject responses to the one or more audio stimuli may comprise an intensity of the one or more audio stimuli that the one or more training subjects detect, hear, and / or detect the one or more audio stimuli. In some embodiments, the untrained or partially untrained machine learning algorithm and / or predictive model may comprise a neural network, support vector machine, random forest, naive Bayes classification algorithm, gradient-boosting algorithm, linear regression algorithm, unsupervised clustering algorithm, supervised clustering algorithm, or any combination thereof. In some embodiments, the trained machine learning algorithm and / or the trained predictive model may be configured to predict and / or determine one or more subjects’ hearing classification with at least about 85% accuracy, specificity, sensitivity, or any combination thereof. In some embodiments, the one or more training subject responses may comprise one or more responses of the one or more training subjects to the one or more audio stimuli provided to a left ear and / or a right ear of the one or more training subjects’ ears. In some embodiments, the one or more audio stimuli may be provided to the one or more training subjects by a computing device. In some embodiments, the computing device may comprise a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, the one or more training subject responses may comprise pressing a surface when the one or more training subjects detect or hear the one or more audio stimuli. In some embodiments, the surface may comprise a surface of the computing device. In some embodiments, the trained machine learning algorithm and / or the trained predictive model may be configured to receive an input of a subject’s one or more responses of detecting, hearing, and / or identifying one or more audio stimuli provided to the subject’s ear. In some cases, the trained machine learning algorithm and / or the trained predictive model may be configured to output a hearing classification of the subject. In some embodiments, the trained machine learning algorithm and / or the trained predictive model may comprise a plurality of trained machine learning algorithms or a plurality of trained predictive models. Provided herein, in some embodiments, is a system configured to train an untrained or partially untrained machine learning algorithm and / or predictive model, comprising: one or more processors and memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions to: receive and / or obtain one or more training subject responses to one or more audio stimuli provided to the one or more training subjects’ ear(s) and corresponding one or more hearing classification of the one or more training subjects; and train the untrained or the partially untrained machine learning algorithm or predictive model with the one or more responses and corresponding hearing classifications thereby generating a trained machine learning algorithm and / or trained predictive model. In some embodiments, the one or more training subject responses to one or more audio stimuli may comprise an intensity of the one or more audio stimuli that the one or more training subjects detect, hear, and / or detect the one or more audio stimuli. In some embodiments, the untrained or the partially untrained machine learning algorithm and / or predictive model may comprise a neural network, support vector machine, random forest, naive Bayes classification algorithm, gradient-boosting algorithm, linear regression algorithm, unsupervised clustering algorithm, supervised clustering algorithm, or any combination thereof. In some embodiments, the trained machine learning algorithm or the trained predictive model may be configured to predict or determine one or more subjects’ hearing classification with at least about 85% accuracy, specificity, sensitivity, or any combination thereof. In some embodiments, the one or more training subject responses may comprise one or more responses of the one or more training subjects to the one or more audio stimuli provided to a left ear and / or a right ear of the one or more training subjects’ ears. In some embodiments, the one or more audio stimuli may be provided to the one or more training subjects by a computing device. In some embodiments, the computing device may comprise a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. In some embodiments, the one or more training subject responses may comprise pressing a surface when the one or more training subjects detect or hear the one or more audio stimuli. In some embodiments, the surface may comprise a surface of the computing device. In some embodiments, the trained machine learning algorithm or the trained predictive model may be configured to receive an input of a subject’s one or more response of detecting or hearing one or more audio stimuli provided to the subject, and where the trained machine learning algorithm and / or the trained predictive model may be configured to output a hearing classification of the subject. In some embodiments, the trained machine learning algorithm and / or the trained predictive model may comprise a plurality of trained machine learning algorithms and / or a plurality of trained predictive models. In some embodiments, the one or more programs may further comprise instructions to generate or update an auditory health profde of the subject based on one or more of the determined hearing classification or the auditory health recommendation provided based thereon.

[0051] In a particular embodiment that may be implemented in conjunction with the systems and methods described above, the machine learning algorithm may be built on a dataset of around 100,000 anonymised audiograms collected in clinical settings using standardised hearing test protocols and calibrated hardware (a particular database includes 110,000 audiograms). Each audiogram of the particular collection contains eight core features—hearing thresholds at 500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz for both ears. A high-quality subset of 500 audiograms was manually annotated by expert audiologists across key diagnostic dimensions, including hearing loss presence, severity, asymmetry, and configuration. These annotations form the clinically validated ground truth and training data used for model development and performance verification.

[0052] In a particular implementation, the system employs a deterministic rule-based algorithm that classifies hearing loss and referral need based on predefined clinical thresholds. This approach enables transparent, interpretable decision-making aligned with audiological standards. To extend the system’s diagnostic capabilities, implementations may also be designed to incorporate advanced machine learning models, such as Random Forests and Gradient Boosting algorithms. These models can utilise the same structured audiogram data, as well as derived features such as band-averaged thresholds an / or asymmetry scores, to enable probabilistic classification and improved accuracy, particularly for borderline or complex cases.

[0053] The model can be validated using both offline analysis and in-app unit testing and is found to achieve 100% agreement with expert-labelled referral decisions.

[0054] Further details of a particular implementation are set out below. Features of the embodiment described below may be implemented in whole or in part in conjunction with the systems described and claimed herein. Data Collection and Preparation

[0055] The dataset utilised for developing a particular embodiment of the Audiogram Interpretation AI System (one implementation of the machine learning algorithm described herein) comprises audiograms collected through a mobile application utilised in primary and community healthcare settings. The audiograms were obtained during clinical hearing examinations performed using calibrated audiometric headphones (in this embodiment SENNHEISER® HD 300 PRO) to standardise acoustic measurement conditions. The data collection yielded over 130,000 audiogram records. After anonymisation to ensure patient privacy and compliance with data governance requirements, approximately 110,000 records met quality and completeness standards and were subsequently retained for model development.

[0056] Each audiogram includes precisely measured hearing thresholds at standard audiometric frequencies—500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz—for both left and right ears, providing eight numerical threshold values per audiogram (four per ear). Audiograms containing incomplete data points, out-of-range thresholds (beyond clinical plausibility thresholds of -20 dB HL to 120 dB HL), or measurement inconsistencies were systematically excluded, ensuring a dataset with robust integrity and consistency. No patient-specific demographic data (such as age, sex, or medical history) was collected, thus ensuring unbiased generalisability across a diverse clinical population. Data Annotation Process

[0057] A representative subset of 500 audiograms was strategically selected using an unsupervised clustering-based methodology to ensure diverse audiometric profiles. This approach combined Affinity Propagation for initial clustering followed by KMeans clustering, resulting in 160 audiometric profile clusters. The audiograms within these clusters were proportionally sampled, prioritising varied configurations such as flat, sloping, notched, cookiebite, and other complex audiometric shapes.

[0058] Each of the 500 selected audiograms underwent manual annotation by three certified audiologists independently, who classified each audiogram based on clearly defined, clinically established criteria: - Hearing Loss Presence: Audiologists classified hearing as "Outside Normal Limits" if any audiometric frequency threshold was 26 dB HL or greater. Audiograms with all thresholds below 26 dB HL were categorised as "Within Normal Limits." Severity Classification: If hearing was classified as outside normal limits, severity was further defined across three frequency bands: low (100-999 Hz), mid (1000-3999 Hz), and high (4000-8000 Hz). Severity was categorised per ISO audiometric standards as Mild (26-39 dB HL), Moderate (40-69 dB HL), Severe (70-94 dB HL), or Profound (>95 dB HL). - Hearing Loss Asymmetry: Defined as clinically significant differences between ears, specifically identified when one ear exhibited at least a 15 dB HL difference from the other ear at one or more consecutive audiometric frequencies, categorised as Left Worse, Right Worse, or Not Asymmetric. - Audiometric Configuration: Classified according to clinically standardised patterns: Flat, Gradually Sloping, Steeply Sloping, Rising, Cookie-bite, Reverse Cookie-bite, Notched, Comer, or Other. - Measurement Variability: Used in cases where audiologists identified potential measurement anomalies due to patient response variability or technical issues.

[0059] Following independent annotation, a senior audiologist reviewed the labels from all annotators, resolving any discrepancies through consensus-driven review. Inter-rater reliability metrics were calculated using Fleiss-Kappa statistics, achieving high agreement scores, especially regarding critical referral decisions, indicating robust annotation quality. Machine Learning Algorithm Implementation

[0060] Example Deterministic Rule-Based Model

[0061] An example implementation of the Audiogram Interpretation AI System employs a deterministic rule-based algorithm, explicitly defined by clinical audiometric thresholds aligned with standard audiology practice. This rule-based logic serves as the primary method for audiogram classification, referral decision-making, and severity assessment. Specifically, the model operates directly on the audiometric threshold data collected at four standard frequencies—500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz—for both left and right ears, totalling eight distinct predictor inputs per audiogram.

[0062] The classification logic within the deterministic model functions as follows:

[0063] Referral Classification:

[0064] For each ear independently, the system evaluates hearing thresholds at the four core frequencies. If any frequency threshold is equal to or exceeds 26 dB HL, the model categorically identifies that ear as being "Outside Normal Limits," thereby flagging it as requiring clinical specialist referral. Conversely, if all thresholds are below 26 dB HL, the ear is classified as "Within Normal Limits," and no referral is indicated.

[0065] Severity Classification:

[0066] For ears flagged as "Outside Normal Limits," the system calculates average hearing thresholds separately across three clinically recognised frequency bands: Low Frequency Band (500 Hz): Threshold at 500 Hz alone. Mid Frequency Band (1000-2000 Hz): Averaged thresholds at 1000 Hz and 2000 Hz. High Frequency Band (4000 Hz): Threshold at 4000 Hz alone.

[0067] Each frequency band's averaged threshold is categorised into one of four severity levels as per ISO audiometry standards: - Mild: 26-39 dB HL Moderate: 40-69 dB HL Severe: 70-94 dB HL Profound: >95 dB HL

[0068] Severity metrics generated by the rule-based model allow clinicians to rapidly assess the audiometric profde across frequency bands, guiding more precise clinical intervention planning.

[0069] Hearing Loss Asymmetry:

[0070] The rule-based model also explicitly assesses inter-ear asymmetry, calculated as the absolute difference in threshold values between the left and right ears at each corresponding audiometric frequency (500 Hz, 1000 Hz, 2000 Hz, 4000 Hz). Clinically significant asymmetry is defined when one or more adjacent frequency thresholds differ by 15 dB HL or greater, clearly indicating a potentially pathological inter-ear hearing disparity. The model then explicitly classifies asymmetry direction as either "Left Ear Worse," "Right Ear Worse," or "Not Asymmetric."

[0071] Audiometric Configuration and Measurement Variability:

[0072] In one embodiment, audiometric configuration classification and measurement variability detection rely exclusively on manual audiologist annotations and are not automatically determined by the deterministic rule-based model. However, these annotations support ongoing model verification and future algorithmic enhancements.

[0073] This deterministic model approach offers critical advantages, including transparency, clinical interpretability, and reliability—essential characteristics within healthcare diagnostic contexts. The model’s explicit logic enables clear and accountable decision-making, aligning directly with well-established clinical audiometric standards. Advanced Machine Learning Implementation

[0074] Given the structured numeric audiogram data, more advanced iterations of the Audiogram Interpretation AI System are highly amenable to sophisticated machine learning techniques, specifically tree-based ensemble methods. These algorithms complement the deterministic approach by enhancing diagnostic accuracy, especially in ambiguous audiometric cases.

[0075] Gradient Boosted Decision Trees (GBDT):

[0076] GBDT algorithms can be directly trained on the eight audiometric thresholds per patient, alongside derived numeric features such as frequency band averages, asymmetry metrics, and potentially calculated slopes between frequencies. These models iteratively build multiple decision trees, each progressively refining predictions based on the residual errors from previous iterations. GBDT can thus effectively capture subtle audiometric patterns, frequency interactions, and non-linear threshold relationships, providing probabilistic outputs for referral urgency and severity confidence levels.

[0077] Random Forest Classifiers:

[0078] Random Forest models are ensemble classifiers composed of multiple decision trees independently trained on bootstrapped samples of audiogram data. These trees make classification decisions based on audiometric thresholds, band averages, asymmetry scores, and derived metrics. The resulting ensemble output, determined through majority voting, significantly reduces the risk of overfitting and enhances model robustness. Random Forests also facilitate transparent clinical interpretation through the calculation of feature importances, highlighting audiometric features most predictive of specific hearing loss patterns.

[0079] Hybrid Rule-Based and Machine Learning Model:

[0080] A further implementation implements a hybrid model architecture. Initially, the deterministic rule-based logic identifies clearly defined clinical cases—those distinctly within normal limits or presenting severe hearing loss requiring immediate referral. Subsequently, audiograms classified as ambiguous or borderline by the deterministic rules are processed by a supplementary machine-learning component (Gradient Boosting or Random Forest). This secondary model utilises the same audiometric features but applies probabilistic reasoning to resolve clinical uncertainties, further improving diagnostic specificity and clinical utility. Model Validation

[0081] The Audiogram Interpretation AI System underwent validation through a dual-phase procedure, combining offline statistical evaluation with real-time integrated application testing:

[0082] Offline Statistical Validation:

[0083] Validation was conducted using the 500 audiogram subset manually annotated by audiologists, whose consensus-driven classifications served as the definitive reference (ground truth). The system’s deterministic logic predictions were statistically evaluated against these consensus labels using Python-based analytical tools, explicitly calculating accuracy, sensitivity, specificity, and confusion matrices for referral classifications. In this validation, the deterministic model achieved 100% classification accuracy for referral decision-making, validating the alignment of the rule-based system’s clinical thresholds with expert audiologist annotations. In-App Integration and Unit Testing:

[0084] To ensure consistent translation of the model logic from the Python validation environment to the mobile application, extensive unit tests were embedded directly within the mobile application development cycle. Anonymised audiogram data was processed in real-time using the mobile app’s embedded deterministic logic, automatically verifying output classifications against audiologist consensus labels.

[0085] Further example validation strategies include incorporating longitudinal patient outcomes, such as specialist diagnosis verification or patient follow-up records, further refining predictive accuracy and enhancing clinical validation. Robust continuous monitoring practices, including statistical data drift detection, can proactively identify shifts in audiogram data distributions, enabling model recalibration and retraining to maintain performance accuracy over time. Hearing Measurement Techniques

[0086] Hearing measurement can take various forms. Hearing measurements may include subjective and / or objective test measures. Subjective test measures may comprise providing an audio stimulus and waiting for a subject to respond (e.g., by pressing and / or indicating to an observer) that the subject detected, heard, and / or identified the audio stimulus. In some cases, subjective test measures may be heavily influenced by psychometric factors. By contrast, objective hearing test measures may obtain empirical information that is not influenced by psychometric factors and does not rely on human interpretation.

[0087] One example of a hearing measurement technique is pure-tone audiometry used to assess hearing sensitivity. During a pure-tone audiometry test, a subject may wear headphones or inserts earphones into a left and / or right ear of the subject. The subject may listen to pure tone sounds at various frequencies and / or frequency bands (pitch) and intensities (loudness). As will be understood by those skilled in the art, the term “pure tone” refers to a tone with a purely sinusoidal waveform. The individual may respond when they hear, detect, and / or identify a sound by raising their hand or pressing a surface (e.g., a physical and / or software user interface button) that is recorded as a response of the subject to the audio stimulus. Through the course of examination an examiner may record the softest sound (threshold) the person can hear at each frequency and / or frequency band of the audio stimuli provided to the subject.

[0088] Pure-tone audiometry may comprise an air conduction test (e.g., using headphones and / or earbuds to deliver the audio stimuli) and / or a bone conduction (using a bone oscillator to deliver the audio stimuli) test. Air conduction thresholds may measure overall hearing sensitivity at one or more frequencies and / or one or more frequency bands, while bone conduction thresholds may determine one or more types of hearing loss (e.g., conductive, sensorineural, or mixed).

[0089] Speech audiometry may assess the ability to hear and understand speech. During speech audiometry, a subject may be provided one or more audio stimuli of recorded speech at different intensities and may repeat words or sentences. Speech audiometry may help determine speech discrimination ability of a subject and / or the impact of hearing loss on communication for the subject.

[0090] Tympanometry may measure the mobility of the eardrum and / or the function of the middle ear. Tympanometry may help identify middle ear conditions e.g., glue ear, otitis media with effusion, and / or eardrum abnormalities (e.g., tympanic membrane perforation) that may contribute to hearing loss.

[0091] Otoacoustic emissions (OAE) testing may assess the function of the outer hair cells of the cochlear in the inner ear of a subject. During OAE, a small probe may be placed in the subjects’ ear canal to directly deliver one or more audio stimuli to the subject’s ear drum and / or corresponding ear anatomy (e.g., the inner ear conductive bones and / or the cochlea). Healthy inner ear cells may produce faint sounds in response to the one or more audio stimuli, which can be detected and measured to determine the function of the outer hair cells of a subject’s cochlea. Absent or weak OAEs may indicate sensorineural hearing loss. OAE testing may be used for newborn subjects hearing tests and / or screening.

[0092] Auditory brainstem response (ABR) is a neurophysiological test that may measure a subject’s neurological response to one or more audio stimuli. ABR may be used for newborn subjects and / or pediatric hearing assessment, and in cases where behavioral audiometry is challenging (e.g., infants and / or individuals with cognitive impairments).

[0093] An audiogram, described elsewhere herein, may comprise a graphical display of hearing intensity thresholds across one or more frequencies and / or one or more frequency bands of audio stimuli provided a subject, determined by the systems and / or methods described elsewhere herein. In some instances, the hearing threshold may be determined and / or obtained from pure tone audiometry. An audiogram can be measured, detected, and / or observed by manual audiometry and / or automated audiometry. In some cases, hearing intensity thresholds may be obtained and / or determined using the Hughston-Westlake method of threshold determination.

[0094] In some cases, the systems and / or methods described elsewhere herein may be configured to determine one or more audiometric configurations of a subject. Audiometric configurations may comprise patterns of hearing loss that may be observed, determined, and / or analyzed from an audiogram. Audiometric configurations may be utilized to screen, determine, and / or diagnose the type and extent of hearing loss of a subject. For example, a flat hearing loss configuration and / or hearing classification may comprise uniform hearing loss indicated by an increase in intensity of the detected and / or identified audio stimuli provided to a subject across one or more frequencies and / or one or more frequency bands above a threshold intensity or limit of normal hearing. In some cases, the increase in intensity may comprise less than about a 5dB rise or fall per octave of the one or more audio stimuli at the one or more frequencies and / or one or more frequency bands. In some cases, a sloping hearing loss configuration and / or hearing classification may be characterized by normal hearing intensity thresholds (or limits), described elsewhere herein, hearing at low frequencies (e.g., about 1Hz to about 1000Hz) and poorer hearing at high frequencies (e.g., about 4000Hz or more). In some cases, sloping hearing loss configuration and / or hearing classification may comprise a sloping intensity across the one or more audio stimuli frequency and / or frequency bands assessed. In some cases, the threshold intensity indicative of the sloping hearing loss configuration and / or hearing classification may comprise a threshold intensity of detected and / or identified one or more stimuli across the one or more frequencies and / or one or more frequency bands provided to the subject ear increases 25 dB or more per octave of the one or more audio stimuli. In some cases, a rising hearing loss configuration and / or hearing classification may comprise hearing intensity thresholds within normal hearing at higher frequencies (e.g., about 4000Hz or more) compared to lower frequencies (e.g., about 1Hz to about 1000Hz) of the one or more audio stimuli provided to a subject. In some cases, a rising hearing loss configuration and / or hearing classification may suggest conductive hearing issues of a subject. In some instances, the rising hearing loss configured and / or hearing classification may comprise hearing intensity thresholds with at least about 5 dB decrease in the hearing intensity threshold per octave of a subject in response to one or more audio stimuli at one or more frequencies and / or one or more frequency bands. In some cases, a notched hearing loss configuration and / or hearing classification may comprise one or more frequencies and / or one or more frequency ranges with a hearing intensity threshold indicating hearing loss. In some cases, in a notched hearing loss configuration and / or hearing classification, the one or more frequencies and / or one or more frequency bands may comprise a frequency within the range of about 3,000 Hz to about 6,000 Hz. In some cases, in a notched hearing loss configuration and / or hearing classification there may be at least about a 20dB increase in hearing intensity threshold above a normal hearing intensity threshold at a frequency with a normal hearing intensity threshold at an adjacent one or more frequencies. In some cases, a cookie-bite or U-shaped configuration and / or hearing classification may comprise an increase in hearing intensity threshold in the mid-frequencies (e.g., about 1000Hz to about 4000Hz) with normal hearing thresholds at low frequencies (e.g., about 1Hz to about 1000Hz) and high frequencies (e.g., about 4000Hz or more). In some cases, a cookie-bite or U-shaped configured and / or hearing classification may comprise an increase in hearing intensity threshold in the mid frequencies with normal hearing intensity thresholds at low and high frequencies. In some cases, a comer hearing loss configuration and / or hearing classification may comprise normal low-frequency hearing intensity thresholds at about 1Hz to about 1000Hz with severe to profound loss indicated by increased hearing intensity thresholds at higher frequencies (e.g., about 4000Hz or more). Computing Device Application

[0095] Provided herein, in some embodiments, is an application (e.g., a mobile application) configured to determine, classify, and / or screen one or more subjects’ auditory health. In some cases, the application may utilize one or more machine learning algorithms and / or one or more predictive models, described elsewhere herein, to determine, classify, and / or screen one or more subjects’ hearing and determine one or more hearing classifications. In some cases, the application in response to the determined one or more hearing classifications may provide an auditory health referral to a subject. In some cases, the application may be operated on a computing device, as described elsewhere herein. In some instances, the application may measure and / or assess a subject’s hearing by providing one or more audio stimuli to a subject and measuring one or more responses of the subject to the one or more audio stimuli. In some cases, the one or more responses of the subject may be visualized and / or provided a user of the application as an audiogram. FIG. 1 shows the launch of the hearing test conducted by the application. A user may navigate to the hearing test and may hand the device to a subject after filling in settings and / or information for the subject and / or the hearing test. In some cases, the application may comprise a user interface view 101 that shows one or more otoscopy images of the ear previously acquired by the computing device and / or by another user and / or clinician and / or physician. In some cases, the one or more otoscopy images may inform the settings and / or information inputted into the application ahead of conducting the hearing test with the application. In some cases, the user interface view 101 may display atop otoscopy image 130 on the user interface view application screen for the left ear. In some cases, the user interface view 101 may display a bottom otoscopy image 132 on the user interface view for a right ear. In some embodiments, the user interface view 102 may display a text-based menu user interface object 128 with selection options that a user can select to indicate which headphones and / or ear buds will be used for the hearing test. To advance from the user interface view 102 a user may tap and / or select a button object 126 e.g., displaying the text of “next”. In some cases, upon tapping and / or selecting a button object 126 a user interface view 103 may be display that provides instructions and / or an alert to a subject that one or more audio stimuli, described elsewhere herein, will be played in their ear to begin the hearing test. In some cases, the subject and / or user of the application may tap and / or select the button 126 on the user interface view 103 to then commence the hearing test. In some cases, user interface view 104 may then display instructions 122 requesting that the subject tap a surface of the computing device at a user interface object 124 when subject hears one or more audio stimuli. Subject name, date of birth, and / or reference number(s) may be input into the app for user profile creation ahead of navigating to the hearing test user interface views (101,102, 103, and / or 104). Each subject may be associated with a business and user. A customer who has purchased the system and / or the application may be known as a “tenant.” A user may be trained to use the system and / or application by undergoing training and certification. A subject may be the end beneficiary of the system and / or application. A user may use the system and / or application to perform otoscopy, determine if wax removal is required, remove wax using micro suction, perform a hearing test to determine if there is a hearing weakness, or any combination thereof. The application may be configured to capture clinical data captured for subjects e.g., medical history, a consent form, otoscopy images and / or videos, hearing test results, hearing classification, or any combination thereof.

[0096] In some cases, the application may provide an auditory health recommendation to a subject based on the hearing classification, described elsewhere herein. In some instances, the referral may be provided to a subject by the application from a non-ear specialist, such as a pharmacy organization to refer the subject to a private audiology specialist clinic for further clinical evaluation and / or assessment. The referral may be made to the audiology clinic if the non-specialist user operating the application and / or the systems, described elsewhere herein, suspects a hearing weakness that warrants further investigation. The referral may be made to the audiology clinic, where the audiology clinic may then view and analyze the subject’s medical record held e.g., a subject’s medical record held in an electronic medical record system. The subject’s medical record held by the electronic medical record system may comprise subject demographic information, medical history, otoscopy images and / or videos, audiogram data, additional comments made by the electronic medical record system’s clinical review team, comments made by ear, nose, and throat (ENT) surgeons, comments made by a pharmacist, or any combination thereof.

[0097] The images shown in 130 and 132 may be displayed on a mobile device application. In some cases, the application may interface with an otoscopy system e.g., a mobile otoscopy system. The application may be compatible with one or more types and / or brands of mobile phones and corresponding operating systems e.g., iPhone OS and / or Android OS. The application may be utilized to conduct and / or perform otoscopic imaging of a subject’s ear anatomical features as well as administer and / or assess a subject’s auditory health, as described elsewhere herein. In some cases, the images acquired by otoscopy may comprise images of a subject’s eardrum and / or ear canal. In some cases, the images acquired by otoscopy may comprise magnified images of a subject’s eardrum and / or ear canal. In some cases, the images acquired by otoscopy may be acquired and / or detected while a visible light illumination source is provided to the subject’s ear anatomical features of e.g., the subject’s ear drum and / or ear canal.

[0098] In some cases, a user (e.g., a clinician) of the application and / or systems, described elsewhere herein, may visualize the subject’s ear canal and / or eardrum to note, measure, and / or screen the subject’s eardrum and / or ear canal and note any contraindications that may influence the result and / or hearing classification determined from the hearing test ahead of performing the hearing test. For example, if a subject has an active ear infection, then a user of the application may be advised against conducting the hearing test. In some cases, if there is a condition that may impact the reliability of the hearing test such as wax impaction of the ear canal, a user of the systems and / or application, described elsewhere herein, may be prompted to suction the ear wax ahead of performing and / or proceeding with administering a hearing test. A user may be encouraged to acquire and / or record one or more otoscopic images and / or video of a subject’s ear anatomical features in the subject’s medical record. In some cases, a user may skip acquiring and / or recording one or more otoscopic images and / or video of subject’s ear anatomical features ahead of administering a hearing test to the subject. In some cases, the one or more otoscopic images and / or videos may be considered and / or processed in combination with the one or more responses of the subject to the one or more audio stimuli to determine a hearing classification of the subject. In some instances, the one or more otoscopic images and / or videos may not be considered and / or processed in combination with the one or more responses of the subject to the one or more audio stimuli to determine a hearing classification of the subject.

[0099] In some cases, the one or more audio stimuli provided to a subject during the hearing test, described elsewhere herein, may be administered and / or provided to a subject with the use of a headphone and / or earbuds. In some cases, the headphones and / or earbuds may comprise wired and / or wireless headphones and / or ear buds. For example, the one or more audio stimuli provided to a subject during the hearing test may be administered and / or provided to a subject with HD300 Pro headphones, ADAPT 360 wired headphones, DD65 headphones, DD65v2 headphones, or any combination thereof. In some cases, the headphones and / or earbuds may comprise a feature of a reference equivalent sound pressure level measurement, which provides a correction factor applied to hearing test output at the one or more frequencies to reduce measurement uncertainty and error. In some cases, the application may administer and / or provide the hearing test to the subject without designating a headphone and / or ear buds to deliver the one or more audio stimulation to the subject during the hearing test.

[0100] The one or more audio stimuli described elsewhere herein may comprise an audio tone stimulus of between about 1 and about 3 seconds. After the one or more audio stimuli are provided to the subject, there may be a period of silence (e.g., a response window), during which the subject may tap on the surface of the computing device once the subject has detected and / or identified the one or more audio stimuli. In some cases, after the computing device and / or application receive the subject’s response to detecting and / or identifying the one or more stimuli, there may be a period of time (e.g., the inter stimulus interval) between about 1 second and about 3 seconds before the next audio stimulus is provided to the subject. Each stimulus has an associated frequency and intensity. For example, each stimulus or consecutive stimuli may have different frequencies and / or intensities. For example, the computing device and / or application may cycle through, for each of a plurality of frequencies, stimuli having different intensities - an example sequence of stimuli for a particular frequency is shown in FIG. 12.

[0101] In some cases, the application may comprise a user interface view, as shown in FIGS. 2A, 3A-3C, and 14A-14D, that displays a hearing level audiogram 204 corresponding to subject’s one or more responses to the intensity one or more audio stimuli, described elsewhere herein. In some instances, the user interface view of the application, as shown in FIGS. 2A, and 3A-3C, may comprise a text and / or object indicator of a determined status and / or hearing classification of a right ear 200 and / or a left ear 202 of the subject. In some cases, the text and / or object indictor of the determined status and / or hearing classification of the right ear 200 and / or the left ear 202 may provide a graphical object (e.g., a check and / or an alert indicator) to indicate the status and / or hearing classification. In some cases, the status and / or the hearing classification may comprise a status of within normal limits, outside of normal limits, or unable to determine / inconclusive. In some cases, the user interface view may display detailed information 201 comprising information explaining the status and / or hearing classification. For example, the detailed information 201 may display text describing a range of the determined hearing threshold for each ear in comparison to a normal hearing standard, a range of determined and / or detected hearing thresholds outside of normal hearing standards, and / or an inconclusive reading. In some cases, the hearing level audiogram 204 may display a hearing level graph line for a subject’s left ear 206 and / or a hearing level graph line 208 for a subject’s right ear at the one or more frequencies and / or one or more frequency bands of the one or more stimuli provided to the subject during the hearing test. In some cases, the user interface view may comprise a graph legend 228 comprising symbolic representation and description of the graph object markers on the hearing level audiogram 204 indicating and / or representing the subject’s one or more responses to the one or more audio stimuli at one or more frequencies and / or one or more frequency bands for the hearing level graph line of the left ear 206 and / or the right ear 208 of a subject. The graph legend may also display objects indicating an inconclusive response at the one or more audio stimuli frequency and / or one or more audio stimuli frequency bands for a subject’s right ear and / or left ear. In some cases, the user interface view may comprise a button 126 that when pressed and / or tapped may display a user interface view, as shown in FIG. 2B. A subject and / or user of the application may navigate from the user interface view displaying the hearing level audiogram 204 shown in FIGS. 2A, 3A-3C, and 14A-14D to the common audiogram trend information, as shown in FIGS. 2B and 4B, by scrolling down on the user interface view shown in FIGS. 2A, 3A-3C, and 14A-14D and clicking, interact with, and / or tapping a collapsible user interface object (401, 402) to expand the common audiogram trend information (404, 408, 410, 406). In some cases, a subject and / or user of the application may then dismiss and / or hide the common audiogram trend information by swiping down on the expanded collapsible user interface menu and / or tapping and / or clicking on the collapsible user interface object (401, 402).

[0102] In some cases, the common audiogram trend information (404, 408, 410, 406), may comprise a display of one or more common audiogram trends, as shown in FIGS. 4A and 5A. In some cases, the one or more common audiogram trends may comprise display text and or objects. In some cases, each common audiogram trend of the one or more common audiogram trends may comprise a heading of the name of the common audiogram trend 404, a corresponding audiogram 406 with a hearing level graph line for the corresponding common audiogram trend for a left 408 and / or right ears 410, a user interface object to provide detailed information that pertains to the common audiogram trend 405, or any combination thereof. In some cases, a user and / or a subject may swipe and / or scroll left and right once the common audiogram trend menu is expanded to view and interact with the one or more common audiogram trends, as shown in FIGS. 4A and 5A. In some cases, clicking and / or tapping on the user interface object to provide detailed information pertaining to the common audiogram trend 405 may direct a user and / or a subject to a detailed user interface view further information of the common audiogram trend, as show in FIGS. 2B and 4B.

[0103] In some cases, the application may comprise a user interface view, as shown in FIGS. 2B and 4B, that may display a common audiogram trend 210 and related information e.g., for flat hearing loss, asymmetric hearing loss, sloping high-frequency hearing loss, reverse sloping hearing loss, notch hearing loss, steeply sloping hearing loss, cookie-bite hearing loss, or any combination thereof. In some cases, the user interface view showing the common audiogram trend and related information may be compared with the subject’s audiogram. In some cases, the user interface view may comprise a title 212 of the common audiogram trend e.g., flat hearing, a representative audiogram 210 illustrating the common audiogram trend with a left ear hearing level graph line 216 and / or a right ear hearing level graph line 214, description of the hearing loss of the common audiogram trend 218, clinical advice pertaining to the common audiogram trend 220, advice to the subject for the common audiogram trend 222, reference literature support the information displayed for the common audiogram trend 414,or any combination thereof.

[0104] In some cases, the application may comprise a user interface view, as shown in FIGS. 2C, and 14D that may display an overlay 224 of common audiogram trends onto the subject's hearing level audiogram 204. In some cases, a user and / or a subject may select and / or tap a common audiogram trend from the common audiogram trend menu (404, 408, 410, 406), to then display a corresponding overlay on the subject’s hearing level audiogram 204. In some cases, the overlay may show the approximate and expected range of hearing test results based on the common audiogram trend selected seen in clinical practice. In some cases, the user interface view may display a hearing level audiogram 204 comprising a hearing level audiogram graph line for a right ear 208 and / or a left ear 206 of a subject with an overlay 224 of the common audiogram trend.

[0105] FIGS. 3A-3C show various examples of hearing level audiograms 204 and corresponding hearing classification and / or status of subject’s left 202 and / or right ears 200.

[0106] FIGS. 3A and 14A shows an example of an audiogram within normal limits. In some cases, the one or more machine learning algorithms and / or one or more predictive models analyze and / or process the one or more subject hearing intensity responses to the one or more audio stimuli at the one or more frequencies and / or frequency bands to determine a subject’s corresponding hearing classifications and / or status of subject’s left and right ears. In some cases, values of the one or more subject hearing intensity responses between about 0 dB and about 25 dB may indicate that a subject’s hearing is within normal limits. In some cases, in response to a finding of the subject’s hearing is within normal limits a color of the user interface banner 203 may be adjusted to e.g., blue. In some cases, the user interface view of the application may display a recommendation to the user 230 provided by the one or more machine learning algorithms and / or one or more predictive models analyzing and / or processing the hearing threshold audiogram, as shown in FIG. 14A.

[0107] FIGS. 3B and 14B show an example of an audiogram where one or more of the subject’s hearing intensity responses to the one or more audio stimuli at the one or more frequencies and / or one or more frequency bands are outside of normal limits. In some cases, the one or more machine learning algorithms and / or one or more predictive models may designate and / or determine a hearing classification of outside of normal limits when the one or more hearing intensity responses of a subject’s ear comprise one or more intensity values of at least about 25 dB. In some cases, the user interface may display a color of a user interface banner 203 (e.g., purple) indicating that the subject’s hearing intensity responses are outside of normal limits. In some cases, a determination of a subject’s hearing intensity responses outside of normal limits may be associated with caution and recommendation of an auditory health referral to the subject. In some cases, the user interface view of the application may display a recommendation to the user 230 provided by the one or more machine learning algorithms and / or one or more predictive models analyzing and / or processing the hearing threshold audiogram, as shown in FIG. 14B. The user can mark that specific ear as “recommended for detailed hearing assessment.” In some cases, a detailed hearing assessment may comprise a formal pure tone audiometry analysis. A detailed hearing assessment may comprise air conduction and / or bone conduction hearing tests.

[0108] In some embodiments, the application user interface may display an indicator of the determined status and / or hearing classification of the right ear 200 and / or the left ear 202 of a subject as outside of normal limits, e.g., each ear of the subject with a response intensity to the one or more stimuli across the one or more frequencies and / or frequency bands of at least about 25dB, as shown in FIG. 14D. In some cases, the user interface view of the application may display a recommendation to the user 230 (e.g., a recommendation to complete a comprehensive hearing assessment) provided by the one or more machine learning algorithms and / or one or more predictive models analyzing and / or processing the hearing threshold audiogram 204, as shown in FIG. 14D.

[0109] FIGS. 3C and 14C show an example of an audiogram where an ear of a subject has been labeled and / or classified as inconclusive or no response indicated. In some cases, upon a label and / or classification of a subject’s responses to the one or more stimuli at one or more frequencies and / or one or more frequency bands inconclusive or no response indicate, the user interface banner 203 may display a color, e.g., purple. In some cases, the text and / or object indictor of the determined status and / or hearing classification of the right ear 200 may be marked as unable to determine. In some cases, upon a finding of an inconclusive and / or no response status, a user may use their clinical judgment and either retake the hearing test or mark that ear as recommended for detailed hearing assessment. In some cases, the user interface view of the application may display a recommendation to the user 230 provided by the one or more machine learning algorithms and / or one or more predictive models analyzing and / or processing the hearing threshold audiogram, as shown in FIG. 14C.

[0110] In some cases, the user interface view of the application may comprise a display of one or more common audiograms that can be scrolled through, as show in FIG. 4A. A collapsible user interface object (401,402) may reveal and expose tiles of common audiograms that can be scrolled through and interacted with by a user and / or subject. In some cases, clicking, touching, and / or tapping a user interface object 405 e.g., labeled as “find out more”, may open and / or display a user interface view displaying detailed information pertaining to the common audiogram trend associated with the tile where the user interface object is located, as shown in FIGS. 2B and 4B. In some embodiments, the detailed information user interface may comprise detailed text taken from academic sources with descriptions, clinical implementation, functional performance, references from academic sources, or any combination thereof. In some cases, the detailed information user interface may be dismissed and / or closed by pulling the pop-up down or clicking on the background. In some cases, the chevron 401 next to the common audiogram trend label 402 may be clicked to expand the collapsed common audiogram trend menu. In some cases, the common audiogram trend menu may be collapsed in a baseline user interface state. In some cases, the common audiogram trend expanded menu may comprise a heading 402 that display the heading of the section with common audiogram trends. In some cases, each common audiogram trend of the one or more common audiogram trends may comprise a heading 404 with text e.g., “sloping high frequency hearing loss” indicating the common audiogram trend and corresponding audiogram shown below the heading 404. In some cases, the common audiogram trend menu may comprise an audiogram 406 of the common audiogram trend e.g., sloping high frequency hearing loss. In some instances, the common audiogram trend may comprise a hearing level graph line for the left ear 408 and a hearing level graph line for the right ear 410. [OlH] In some cases, the user interface view of the application displaying a hearing level audiogram 204, as shown in FIGS. 2A, 3A-3C, 4A, and 5B, may comprise one or more patient audiogram assessment questions for the right ear 412 and / or the left ear 506. In some cases, the user interface view of the application displaying a hearing level audiogram may comprise a user object, that upon clicking and / or tapping may take a user and / or the subject to a referral user interface view and / or to the end of the appointment.

[0112] FIG. 4B shows a user interface view of the application comprising detailed information of a common audiogram trend, e.g., flat hearing loss. In some cases, detailed information of the common audiogram trend user interface view may display a title of the common audiogram trend for which information is displayed 212, one or more frequency sample audiograms of common audiogram trends (210, 214, 216), a description of the trend 218, clinical advice 220, subject advice 222, literature based references 414 (supporting the description of the trend, clinical advice, and / or subject advice), or any combination thereof. In some cases, the subject advice 222 may be similar to a script that health care personnel can adapt to explain the condition to a subject.

[0113] FIG. 5A shows a user interface view of the application where a user and / or the subject can click and / or tap on tiles of one or more common audiogram trends to produce an overlay on the subject's audiogram to show the shape of the common audiogram trend superimposed on the subject’s a hearing level audiogram. The overlay may vary by which common reference audiogram the user and / or the subject selects.

[0114] FIG. 5B shows a user interface view of the application displaying one or more questions (412, 506) that a user may answer to grade a subject’s hearing level audiogram and the corresponding status and / or hearing classification of a left and / or right ear of the subject provided by the one or more machine learning algorithms and / or one or more predictive models. Using the systems and / or methods described elsewhere herein, a user of the computing device and application who has a referral feature of the application active may be asked to assess the subject’s ears to determine if the subject’s ears hearing classification and / or status is characterized as “within normal limits”, “recommended for detailed hearing assessment”, or “N / A.” A determination of “N / A” may indicate that the hearing test could not be performed on the subject’s ear. The referral feature may be activated at the subject level by the system. When a referral is sent, it may notify the health care provider and / or physician recipient via an email. The recipient can access the cloud system and view the referral. The user can be a doctor, attending physician, specialist, surgeon, pharmacist, healthcare assistant, physical therapist, ear care practitioner, other type(s) of healthcare provider, or any combination thereof. In some cases, the user interface of the application may display audiogram assessment questions for the right ear 412. In some instances, the user interface of the application may display audiogram assessment questions for the left ear 506. In some cases, the user interface of the application may display publications and supporting literature references 508 for the audiogram assessment. The user may need to select options for both assessment questions before the next button 126 becomes active to select.

[0115] FIG. 6A shows a clinical summary user interface view of the application that may be displayed if a user selects that the subject’s hearing classification of the left and / or right ears are "within normal limits" or “N / A.” In some cases, a user may be able to end a hearing test and / or hearing assessment of the subject at the clinical summary user interface view of the application. In some cases, the clinical summary user interface view of the application may comprise a heading 602 indicating that the user interface view is the clinical summary section. In some instances, the clinical summary user interface view may comprise the subject’s name and birthdate 604. In some cases, the clinical summary user interface view may comprise a menu of procedures completed 606 that a user may select. In some cases, the clinical summary user interface view of the application may comprise a user object (e.g., a button) 608 that a user may select to indicate that otoscopy, ear wax removal, hearing screening check, or any combination thereof have been completed. In some cases, the clinical summary user interface view of the application may comprise a user interface object (e.g., a button with text “finish appointment”) that a user may tap and / or select to complete an appointment e.g., if the subject’s hearing level audiogram is within normal limits.

[0116] FIGS. 6B and 8 shows a clinical summary referral user interface view 105 of the application that may be displayed if a user / clinician has selected answers to the one or more questions (412, 506) that the left and / or right ear of a subject are "recommended for detailed hearing assessment.” In some cases, a user may be provided a user interface object 614 if the user selects answers to the one or more questions (412, 506) pertaining to the subject’s ear as “recommended for detailed hearing assessment.” In some cases, the user interface object may display text to refer the subject. In some cases, the clinical summary referral user interface may comprise a user interface object with text displayed to “skip referral.” In some cases, the clinical summary referral user interface view may comprise a header 602 that indicates that the user interface view is the clinical summary referral user interface view. In some instances, the clinical summary referral user interface view may comprise a user interface object 604 with text of the name and date of birth of the subject. In some cases, the clinical summary referral user interface view may comprise a menu 606 displaying the clinical procedures that have been completed for the subject. In some cases, the clinical summary referral user interface view may comprise one or more user interface objects 608 that may be selected to indicate a procedure of otoscopy, ear wax removal, hearing screening check, or any combination thereof, that have been completed. In some instances, the clinical summary referral user interface view may display information 612 pertaining to referring the subject to a doctor, attending physician, specialist, surgeon, pharmacist, healthcare assistant, physical therapist, or any combination thereof, if one or more ears of the subject are classified and / or determined to have a hearing classification that is abnormal. In some cases, the clinical summary referral user interface view may comprise a user object (e.g., a button) 614 for referring a subject to a doctor, attending physician, specialist, surgeon, pharmacist, healthcare assistant, physical therapist, or any combination thereof. In some cases, the clinical summary referral user interface view may comprise a user object 616 to skip providing a referral for the subject. In some instances, if the “skip referral” button is selected, the application may navigate to user interface view indicating the end of the appointment screen and / or the home screen of the application.

[0117] In some embodiments, the application may comprise a user interface view displaying information that a user collects from a subject 106 when completing a referral, as shown in FIGS. 7A and 8. In some cases, a user may obtain, acquire, and / or collect the subject’s email address, phone number, postal code, consent from the subject to share relevant subject information, or any combination thereof. In some embodiments, the user interface view may display instructions for referral 702. In some embodiments, the user interface view may display one or more confirmation boxes 704 that indicate that the information provided by the subject is verified and / or explained to the subject. In some embodiments, the user interface view may display one or more input fields 706 where a subject may provide and / or input their email address, phone number, postal code, or any combination thereof. In some embodiments, a signature may or may not be collected when the subject is providing consent.

[0118] In some embodiments, the application may comprise a user interface view displaying a list of providers that the subject’s referral may be sent to 107, as shown in FIGS. 7B and 8. In some embodiments, the user interface view may display a list of providers 708 that may receive the subject’s referral. In some embodiments, the user may present the subject with a list of organizations that has been downloaded onto the device ahead of time. The subject may decide on the organization and may select and / or provide instructions to the user to select the health care provider and / or organization they would like their referral sent to by selecting the provider with the user interface object. In some embodiments, the referral may be created if the application and / or the computing device running the application is not connected to the internet.

[0119] In some embodiments, the application may comprise a synchronization user interface view, as shown in FIG. 7C, that may be displayed if the computing device and / or application are not connected to the internet to, e.g., transmit a referral to an auditory health care provider. In some cases, the synchronization user interface may display one or more subjects’ referrals (710, 712) placed in a Sync queue. In some embodiments, once the computing device and / or the application are connected to the internet, the one or more subjects’ referrals (710, 712) may be sent to providers when the device connects to the internet. In some embodiments, the synchronization user interface view may display a subject’s referral and the corresponding fdes 710 that are queued to be sent to a provider. In some embodiments, the synchronization user interface view may display a subject’s uploaded otoscopy results and referral 712 that have finished sending to the provider. In some cases, subject’s referrals are automatically synchronized. In instances where sync has not taken place automatically, a user can force the application to synchronize referrals that are queued.

[0120] In some embodiments, the application may comprise a user interface for accessing referrals, as shown in FIG. 9A. In some cases, the user interface for accessing referrals may be provided on a computer system, a cloud-based platform, a web portal, a personal computing device, a tablet, a smart phone, or any combination thereof. The user interface for accessing referrals may be used to manage receiving referrals and / or initiating follow up appointment scheduling with subjects. Subject details may not be shown to referees until they are selected for contacting. Subject information including contact details, postcode, and subject history may be made visible after the referee organization accepts the referral. The user interface for accessing referrals may have several functionalities: access to a subject’s complete medical history, telemedicine support for clinicians where they have referred a subject record to the system clinical team for specialist advice, device and user management, device configurations, activity analytics, or any combination thereof. In some embodiments, the user interface for accessing referrals may comprise a heading 902 indicating that the user interface is displaying auditory referrals, described elsewhere herein. In some cases, the user interface for accessing referrals may comprise a menu to filter 904 to filter by provider location that has received a referral. In some embodiments, the user interface for accessing referrals may comprise a date range selection 906 configured to limit referrals received within a specified date range. In some embodiments, the user interface for accessing referrals may comprise a referral status menu 908 comprising one or more user interface selection box objects that may filter referrals in the referral list by status, where status may comprise a new referral, a referral that has been accepted, a referral where contact to the subject has been attempted, a referral that has been completed, or any combination thereof. In some embodiments, the user interface for accessing referrals may comprise a list of referrals of one or more subjects received 910. The referral list may comprise information of the date the referral was received, the referral reference number, the subject ID, the referral location, the subject’s name, the subject’s date of birth, referral status, whether the referral is complete, or any combination thereof.

[0121] In some embodiments, the application may comprise a user interface for accessing patient information from a subject’s referral, as shown in FIG. 9B. In some embodiments, the user interface for accessing patient information from a subject’s referral may comprise a heading 912 indicating that the user interface is displaying information of a subject of interest. In some embodiments, the user interface for accessing patient information from a subject’s referral may comprise subject information 914, where the subject information comprises the subject’s date of birth, email, phone number, or any combination thereof. In some embodiments, the user interface for accessing patient information from a subject’s referral may comprise referral status information for the subject 916. In some embodiments, the referral status information may comprise information of whether contact to the subject has been attempted, the referral has been accepted, the referral is completed, or any combination thereof. In some embodiments, the user interface for accessing patient information from a subject’s referral may comprise information about the subject’s appointments 918. In some embodiments, information about the subject’s appointments may comprise the dates of the appointments and the number of appointments. In some embodiments, the user interface for accessing patient information from a subject’s referral may comprise information about the subject’s hearing tests 920 e.g., the date of the hearing test, hearing classifications of a left and / or right ear of the subject, further recommended assessment and follow up, or any combination thereof. In some embodiments, the user interface for accessing patient information from a subject’s referral may comprise a menu to filter 904 to e.g., for provider location that has received a referral. In some embodiments, the user interface for accessing patient information from a subject’s referral may comprise a list of medical history forms for the subject 922 and the corresponding health care personnel, physician, ear nose and throat physician, or combination thereof that has conducted a further examination, test, and / or assessment of the subject.

[0122] In some embodiments, the methods described elsewhere herein may comprise a method for providing an auditory health recommendation to a subject from the perspective of a provider clinician, as shown in FIG. 10. In some embodiments, the methods comprise conducting and / or receiving an indication and / or confirmation that otoscopy imaging and / or micro suction of a subject’s ear has been completed 135. In some embodiments, the results of the otoscopy imaging (e.g., one or more images and / or one or more video clips and / or segments of a subject’s ear) conducted and / or performed on the subject may be uploaded to the cloud 136. In some cases, the results of the otoscopy imaging may be uploaded to the cloud during and / or after the completion of the otoscopy imaging. In some embodiments, the method for providing an auditory health recommendation may comprise conducting a hearing test 137, as described elsewhere herein, e.g., by providing one or more audio stimuli to the subject and determining a hearing classification of the subject from one or more responses of the subject to the one or more audio stimuli. In some embodiments, an application on a computing device, described elsewhere herein, may process and / or analyze the subject’s one or more responses to the one or more audio stimuli to generate and / or determine the hearing classification 138 and determine and / or provide an auditory health recommendation of the subject based on the hearing classification. In some embodiments the audiogram data representing the subject’s one or more responses may be uploaded to a cloud server and / or a remote computing device 139. In some embodiments, the subject may decline the auditory health recommendation determined and / or provided by the systems and / or the application, described elsewhere herein, and / or the systems and / or application may determine that further follow up by a health care personnel for the subject is not needed 140. In some embodiments, the subject may submit information including e.g., name, date of birth, audiogram determined and / or collected by the application and / or system, hearing classifications, or any combination thereof, to a selected referral organization and / or provider 141. In some embodiments, the subject information and the auditory health referral may be uploaded to cloud and sent and / or provided to the referral organization and / or provider 142.

[0123] In some embodiments, the methods, described elsewhere herein, may comprise a method of receiving and / or processing an auditory health referral of a subject, as shown in FIG. 11. In some embodiments, the referee organization and / or provider may receive and / or obtain a subject’s referral by electronic communications (e.g., email) and / or on a corresponding user interface (e.g., on a web portal and / or cloud application platform) 143. In some embodiments, the subject information, described elsewhere herein, may be hidden when the provider and / or organization receives the auditory health referral. In some embodiments, the referee organization and / or provider may accept the referral. In some cases, when the referee organization and / or provide accepts the referral, the subject information, audiogram, and / or hearing classification may be revealed and / or displayed to the referee organization and / or provider 144. In some embodiments, the referee organization and / or provide may review the subject’s information, audiogram, and / or hearing classification 145. In some embodiments, the referee organization and / or provider may contact the subject via electronic communication (e.g., telephone call, email, and / or text-based message) 146. In some embodiments, once a subject is reached, an appointment between the subject and / or the referee organization and / or provider may be made 147. In some embodiments, when contact between the referee organization and / or provider and the subject has been established and the appointment with the subject has been made, the referral may be marked as completed. In some embodiments, a PDF of the subject’s information, audiogram, and / or hearing classification may be downloaded by the referee organization and / or provider and sent to an office for further evaluation. In some embodiments, when the referee organization and / or provider attempts to reach out to the subject and no response from the subject was received and / or contact established, the status of the subject’s referral may change to contact attempted 148.

[0124] In some embodiments, the application may comprise a user interface view of a home screen, as shown in FIG. 13A. In some embodiments, the home screen may comprise a planner displaying the next subject appointment information 1304 for the day including appointment time, and / or appointment type. In some cases, the home screen user interface may comprise a header 1302 displaying the application user’s name. In some embodiments, the home screen user interface may comprise a user interface object (e.g., a button) 1306 to create a new appointment. In some embodiments, the home screen user interface may comprise a user interface object (e.g., a button) 1308 to scan a QR code.

[0125] In some embodiments, the application may comprise a planner list user interface view, as shown in FIG. 13B. In some cases, the planner list user interface view may comprise a header 1310 that indicates that the user interface view displayed is the planner list showing details of the list of appointments for a given day 1312.

[0126] In some embodiments, the application may comprise a calendar user interface view, as shown in FIG. 13C. In some embodiments, the calendar user interface view may display a calendar 1314 in a month, week, and / or day view. In some cases, the calendar user interface view may display an appointment list 1316 for a selected date on the calendar 1314.

[0127] In some embodiments, the application may comprise a subject list user interface view, as shown in FIG. 13D. In some cases, the subject list user interface view may display a search box user interface object 1318, where a user of the application may search for one or more subjects based on subject name, subject number, or a combination thereof. In some cases, the subject list user interface view may comprise a subject list 1320 in result to the searched subjects provided and / or inputted into the search box user interface 1318. In some cases, a user may tap and / or click on a subject of the subject list to view detailed information of the subject.

[0128] In some embodiments, the application may comprise a synchronization user interface view, as shown in FIG. 13E. In some embodiments, the synchronization user interface view may display otoscopic imaging, hearing threshold audiogram data, hearing classification data, subject medical record data, auditory health referral, or any combination thereof, of one or more subjects (710, 712, 1322) to be uploaded to a server and / or remote computing system memory. In some cases, the synchronization user interface view may comprise user interface text objects for a subject’s data comprising information of the subject’s name, date of appointment and / or assessment via otoscopy and / or a hearing test administered, subject’s medical identification information, subject medical data, or any combination thereof. In some cases, the synchronization user interface view may display an indicator 1323 of the status of the synchronization of the subject otoscopic imaging data, hearing threshold audiogram data, hearing classification data, subject medical record data, auditory health referral, or any combination thereof. In some cases, the status of the synchronization may comprise queued 710, finished 712, or failed 1322.

[0129] In some embodiments, the application may comprise a help and / or instructional user interface view, as shown in FIG. 13F. In some cases, the help and / or instructional user interface view may comprise a heading for the instructional materials 1324. In some cases, the help and / or instructional user interface view may comprise a graphical object 1326 of the systems, computing devices, and / or computing systems, described elsewhere herein, that pertain to the instructional or help material and / or information provided in the user interface view displayed is an image of the device. Mobile App Specifications

[0130] In some cases, the methods, systems, and / or application described elsewhere herein may be operated and or used by a user and / or a subject when the computing device is on battery power and / or when the computing device is plugged into a power supply. In some cases, the application may comprise one or more touch screen interfaces. In some cases, an ear simulator may be used for calibration of the methods and / or systems, described elsewhere herein, e.g., as per ANSI / ASA S3.6-2018 standards. In some cases, the ear simulator may be an Artificial Ear e.g., with a coupler compliant with IEC 60318-1. In some cases, the systems and / or computing devices, described elsewhere herein, may require no warm-up time ahead of use. In some cases, the one or more audio stimuli provided by the computing device may comprise a pure audio tone using e.g., a Modified Hughson Westlake algorithm. In some cases, the sound attenuation characteristics of the earbuds and / or heads may be measured in accordance with ISO 4869-1. In some cases, the maximum hearing level settings provided at each test frequency may be about 80 dB. In some cases, the time window for the subject’s one or more response for automated test procedures may be about 2 to about 3 seconds. In some cases, the device systems and / or computing devices, described elsewhere herein, may be calibrated before shipment. In some cases, the computing device and / or systems, described elsewhere herein, may be calibrated in dB HL for known reference equivalent sound pressure levels (RETSPLs) for each audio stimulation transducer (e.g., earbuds and / or headphones, described elsewhere herein) according to ANSI / ASA S3.6-2018. In some cases, a modified Hughson Westlake algorithm may be used or when the one or more audio stimuli are repeated. In some cases, the duration of the initial sound pressure or vibratory force wave of a click and / or duration and rise / fall times of acoustic or vibratory tone-bursts may be as specified in ANSI / ASA S3.6-2018 section 7.5.4. In some cases, the subjective relationship between test signals and reference signals may be that RETSPLs from the manufacturers are used to enable normalized hearing level outputs. In some cases, calibration may be performed in a test lab. Types of Hearing Loss

[0131] Audiometric guidance to be included in the application, described elsewhere herein, may vary. For a subject with hearing thresholds (e.g. detected intensities of one or more audio stimuli) within the normal range (e.g., about 0 dB HL to about 25 dB HL) across all frequencies (e.g., about 250 Hz to about 8000 Hz), the user may provide clinical guidance and / or advice that the subject has no significant hearing impairment. In some cases, if the subject has concerns with their hearing, they may be referred for a detailed hearing assessment. Monitoring hearing health periodically may be recommended.

[0132] For a subject with sloping high-frequency hearing loss, e.g., elevated hearing thresholds (e.g., greater than about 25 dB HL) at higher frequencies (e.g., about 4000 Hz to about 8000 Hz) while thresholds at lower frequencies are within the normal range, the user may tailor clinical guidance and / or advice accordingly. The user may advise the subject that the subject appears to have difficulty hearing high-pitched sounds, which can affect speech understanding. They may further advise that the subject may benefit from hearing aids. They may further advise that this type of hearing loss is common with increased age.

[0133] Reverse sloping / low frequency / rising hearing loss may be determined when a subject has elevated thresholds (e.g., greater than about 25 dB HL) at lower frequencies (e.g., about 250 Hz to about 1000 Hz) while thresholds at higher frequencies are within the normal range. The subject may have difficulty hearing low-pitched sounds. Medical evaluation may be recommended to determine if there is an ear pathology in line with the reverse sloping / low frequency / rising hearing loss. Running the hearing test in a noisy room can create a false low-frequency hearing loss. If the user believes this has occurred, they can run the hearing test again in a quieter room. After medical evaluation and treatment, hearing aids can be discussed with an audiologist.

[0134] Flat hearing loss may be determined when hearing loss is relatively uniform across all frequencies. This causes reduced overall hearing sensitivity, which can affect speech understanding. The subject may benefit from hearing aids.

[0135] Notch hearing loss may be determined when a specific frequency or range shows significantly worse hearing, typically around about 3000 Hz to about 6000 Hz. In this method, at least about 25 dB loss at one frequency with complete or near-complete recovery at adjacent frequencies may be observed. Notch hearing loss can cause difficulty in hearing high-pitched sounds, like consonants, which can affect speech understanding. Such individuals may benefit from hearing aids.

[0136] Steeply sloping hearing loss may be determined when there is a rapid increase in thresholds from low to high frequencies (e.g., greater than about 25 dB HL). Such individuals may have significant difficulty hearing both high-pitched sounds and speech. They may benefit from hearing aids.

[0137] Cookie-bite hearing loss may be determined when there are elevated thresholds (e.g., greater than about 25 dB HL), with a characteristic dip or “bite” in the mid-frequency range (e.g., around about 1000 Hz to about 2000 Hz) while thresholds at low and high frequencies are relatively normal. Such individuals may have difficulty hearing sounds in the mid-frequency range, which can affect speech understanding. Such individuals may benefit from hearing aids.

[0138] Asymmetric hearing loss may be determined where there is a significant difference in hearing ability between a subject’s two ears. Asymmetric hearing loss may be classified as a difference of at least about 15 dB difference in hearing threshold between the two ears of a subject at two or more adjacent frequencies. On some occasions this can indicate a serious underlying medical condition. Subjects with identified asymmetric hearing loss may be recommended for medical evaluation. Asymmetric hearing loss may impact a subject’s ability to hear direction of sounds or take part in conversations, especially in noisy environments. Computer systems

[0139] The present disclosure provides computer systems that are programmed to implement methods and / or application of the disclosure, described elsewhere herein. FIG. 15 shows an example computer system 1601 that is programmed or otherwise configured to assess hearing of a subject. The computer system 1601 can regulate various aspects of the present disclosure, such as, for example, using artificial intelligence to assess a subject’s hearing. The computer system 1601 can be a computing device of a user, and / or a computer system that is remotely located with respect to the electronic device. The computing device can be a mobile electronic device.

[0140] The computer system 1601 includes a central processing unit (CPU, also “processor” and “computer processor” herein) 1605, which can be a single core or multi core processor, or a plurality of processors for parallel processing. The computer system 1601 may comprise memory or memory location 1610 (e.g., random-access memory, read-only memory, and / or flash memory), electronic storage unit 1615 (e.g., hard disk), communication interface 1620 (e.g., network adapter) for communicating with one or more other systems, and peripheral devices 1625, such as cache, other memory, data storage and / or electronic display adapters. The memory 1610, storage unit 1615, interface 1620, peripheral devices 1625, or any combination thereof, may be in communication with the CPU 1605 through a communication bus (solid lines), such as a motherboard. The storage unit 1615 can be a data storage unit (or data repository) for storing data. The computer system 1601 can be operatively coupled to a computer network (“network”) 1630 with the aid of the communication interface 1620. The network 1630 can be the Internet, an internet and / or extranet, or an intranet and / or extranet that is in communication with the Internet. The network 1630, in some cases, may be a telecommunication and / or data network. The network 1630 can include one or more computer servers, which can enable distributed computing, such as cloud computing. The network 1630, in some cases with the aid of the computer system 1601, can implement a peer-to-peer network, which may enable devices coupled to the computer system 1601 to behave as a client or a server.

[0141] The CPU 1605 can execute a sequence of machine-readable instructions, which can be embodied in a program, application, and / or software. The instructions may be stored in a memory location, such as the memory 1610. The instructions can be directed to the CPU 1605, which can subsequently program or otherwise configure the CPU 1605 to implement methods of the present disclosure. Examples of operations performed by the CPU 1605 can include fetch, decode, execute, and writeback.

[0142] The CPU 1605 can be part of a circuit, such as an integrated circuit. One or more other components of the system 1601 can be included in the circuit. In some cases, the circuit may be an application specific integrated circuit (ASIC).

[0143] The storage unit 1615 can store files, such as drivers, libraries and / or saved programs. The storage unit 1615 can store user subject hearing threshold audiogram data, subject medical data, subject name, subject date of birth, subject appointment information, etc. as described elsewhere herein, or any combination thereof. In some cases, the storage unit 1615 may store, e.g., user preferences and user programs. The computer system 1601 in some cases can include one or more additional data storage units that are external to the computer system 1601, such as located on a remote server that is in communication with the computer system 1601 through an intranet or the Internet 1630.

[0144] The computer system 1601 can communicate with one or more remote computer systems through the network 1630. For instance, the computer system 1601 can communicate with a remote computer system of a user (e.g., a mobile phone comprising a mobile application). Examples of remote computer systems include personal computers (e.g., portable PC), slate or tablet PC’s (e.g., Apple® iPad, Samsung® Galaxy Tab), telephones, Smart phones (e.g., Apple® iPhone, Android-enabled device, Blackberry®), and / or personal digital assistants. The user can access the computer system 1601 via the network 1630.

[0145] Methods as described herein can be implemented by way of machine (e.g., computer processor) executable code stored on an electronic storage location of the computer system 1601, such as, for example, on the memory 1610 and / or electronic storage unit 1615. The machine executable and / or machine-readable code can be provided in the form of software and / or an application, as described elsewhere herein. During use, the code can be executed by the processor 1605. In some cases, the code can be retrieved from the storage unit 1615 and stored on the memory 1610 for ready access by the processor 1605. In some situations, the electronic storage unit 1615 can be precluded, and machine-executable instructions are stored on memory 1610.

[0146] The code can be pre-compiled and configured for use with a machine having a processer adapted to execute the code or can be compiled during runtime. The code can be supplied in a programming language that can be selected to enable the code to execute in a pre-compiled or as-compiled fashion.

[0147] Aspects of the systems and methods provided herein, such as the computer system 1601, can be embodied in programming. Various aspects of the technology may be thought of as “products” or “articles of manufacture” typically in the form of machine (or processor) executable code and / or associated data that may be carried on or embodied in a type of machine-readable medium. Machine-executable code can be stored on an electronic storage unit, such as memory (e.g., read-only memory, random-access memory, and / or flash memory) or a hard disk. “Storage” type media can include any or all of the tangible memory of the computers, processors or the like, or associated modules thereof, such as various semiconductor memories, tape drives, disk drives and the like, which may provide non-transitory storage at any time for the software programming. All or portions of the software may at times be communicated through the Internet or various other telecommunication networks. Such communications, for example, may enable loading of the software from one computer or processor into another, for example, from a management server or host computer into the computer platform of an application server. Thus, another type of media that may bear the software elements includes optical, electrical, and electromagnetic waves, such as used across physical interfaces between local devices, through wired and optical landline networks and / or over various air-links. The physical elements that carry such waves, such as wired or wireless links, optical links, or the like, also may be considered as media bearing the software. As used herein, unless restricted to non-transitory, tangible “storage” media, terms such as computer or machine “readable medium” may refer to any medium that participates in providing instructions to a processor for execution.

[0148] Hence, a machine-readable medium, such as computer-executable code, may take many forms, including but not limited to, a tangible storage medium, a carrier wave medium or physical transmission medium. Non-volatile storage media may comprise, for example, optical or magnetic disks, such as any of the storage devices in any computer(s) or the like, such as may be used to implement the databases, etc. shown in the drawings. Volatile storage media may comprise dynamic memory, such as main memory of such a computer platform. Tangible transmission media may comprise coaxial cables, copper wire, and / or fiber optics, comprising the wires that comprise a bus within a computer system. Carrier-wave transmission media may take the form of electric or electromagnetic signals, or acoustic or light waves such as those generated during radio frequency (RF) and infrared (IR) data communications. Common forms of computer-readable media therefore may comprise, for example: a floppy disk, a flexible disk, hard disk, magnetic tape, any other magnetic medium, a CD-ROM, DVD or DVD-ROM, any other optical medium, punch cards paper tape, any other physical storage medium with patterns of holes, a RAM, a ROM, a PROM and EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave transporting data or instructions, cables or links transporting such a carrier wave, or any other medium from which a computer may read programming code and / or data. Many of these forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to a processor for execution.

[0149] The computer system 1601 can include or be in communication with an electronic display 1635 (e.g., of a computing device) that may comprises one or more user interfaces (UI) 1640 for providing, for example, a method for performing a hearing test on a subject. Examples of Ui’s may comprise, without limitation, a graphical user interface (GUI) and / or web-based user interface, described elsewhere herein.

[0150] Methods and / or systems of the present disclosure can be implemented by way of one or more algorithms. An algorithm can be implemented by way of software upon execution by the central processing unit 1605. The algorithm can, for example, analyze a subject’s auditory health. Examples of Machine Learning Methodologies

[0151] As used in this specification and the appended claims, the terms “artificial intelligence,” “artificial intelligence techniques,” “artificial intelligence operation,” and “artificial intelligence algorithm” generally may refer to any system or computational procedure that may take one or more actions that simulate human intelligence processes for enhancing or maximizing a chance of achieving a goal. The term “artificial intelligence” may include “generative modeling,” “machine learning” (ML), or “reinforcement learning” (RL).

[0152] As used in this specification and the appended claims, the terms “machine learning,” “machine learning techniques,” “machine learning operation,” and “machine learning model,” may generally refer to any system or analytical or statistical procedure that may progressively improve computer performance of a task. In some cases, ML may generally involve identifying and recognizing patterns in existing data in order to facilitate making predictions for subsequent data. ML may include a ML model (which may include, for example, a ML algorithm). Machine learning, whether analytical or statistical in nature, may provide deductive or abductive inference based on real or simulated data. The ML model may be a trained model. The ML model may not require training. ML techniques may comprise one or more supervised, semisupervised, self-supervised, or unsupervised ML techniques. For example, an ML model may be a trained model that is trained through supervised learning (e.g., various parameters are determined as weights or scaling factors). ML may comprise one or more of regression analysis, regularization, classification, dimensionality reduction, ensemble learning, meta learning, association rule learning, cluster analysis, anomaly detection, deep learning, or ultra-deep learning. ML may comprise, but is not limited to: k-means, k-means clustering, k-nearest neighbors, learning vector quantization, linear regression, non-linear regression, least squares regression, partial least squares regression, logistic regression, stepwise regression, multivariate adaptive regression splines, ridge regression, principal component regression, least absolute shrinkage and selection operation (LASSO), least angle regression, canonical correlation analysis, factor analysis, independent component analysis, linear discriminant analysis, multidimensional scaling, non-negative matrix factorization, principal components analysis, principal coordinates analysis, projection pursuit, Sammon mapping, t-distributed stochastic neighbor embedding, AdaBoosting, boosting, gradient boosting, bootstrap aggregation, ensemble averaging, decision trees, conditional decision trees, boosted decision trees, gradient boosted decision trees, random forests, stacked generalization, Bayesian networks, Bayesian belief networks, naive Bayes, Gaussian naive Bayes, multinomial naive Bayes, hidden Markov models, hierarchical hidden Markov models, support vector machines, encoders, decoders, autoencoders, stacked auto-encoders, perceptrons, multi-layer perceptrons, artificial neural networks, feedforward neural networks, convolutional neural networks, recurrent neural networks, long short-term memory, deep belief networks, deep Boltzmann machines, deep convolutional neural networks, deep recurrent neural networks, generative adversarial networks, or any combination thereof.

[0153] Methods and / or systems of the disclosure can process, analyze, and / or classify one or more subjects’ hearing threshold audiograms, described elsewhere herein, to determine, classify, diagnose, etc., a subject’s auditory health and / or hearing classification, as described elsewhere herein. In some cases, the processing, analyzing, and / or classifying audiograms and / or one or more features of the audiograms may be conducted by way of one or more machine learning algorithms and / or one or more predictive models with instructions provided with one or more processors as disclosed herein. For example, one or more machine learning algorithms and / or predictive models may process one or more, or two or more features of the hearing threshold audiograms, described elsewhere herein.

[0154] In some cases, the subject's and / or plurality of subjects’ hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with a sensitivity of at least about 70%, at least about 75%, at least about 80%, at least about 85% or at least about 90%.

[0155] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with a sensitivity of up to about 70%, up to about 75%, up to about 80%, up to about 85% or up to about 90%.

[0156] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with a specificity of at least about 70%, at least about 75%, at least about 80%, at least about 85% or at least about 90%.

[0157] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with a specificity of up to about 70%, up to about 75%, up to about 80%, up to about 85% or up to about 90%.

[0158] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with a positive predictive value of at least about 70%, at least about 75%, at least about 80%, at least about 85% or at least about 90%.

[0159] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with a positive predictive value of up to about 70%, up to about 75%, up to about 80%, up to about 85% or up to about 90%.

[0160] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with a negative predictive value of at least about 70%, at least about 75%, at least about 80%, at least about 85% or at least about 90%.

[0161] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with a negative predictive value of up to about 70%, up to about 75%, up to about 80%, up to about 85% or up to about 90%.

[0162] In some cases, the subject's and / or plurality of subjects’ hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with an accuracy of at least about 70%, at least about 75%, at least about 80%, at least about 85% or at least about 90%.

[0163] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with an accuracy of up to about 70%, up to about 75%, up to about 80%, up to about 85% or up to about 90%.

[0164] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with an Area Under the Receiver Operating Characteristic Curve(AUROC) of at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.82, at least about 0.84, at least about 0.86, at least about 0.88, or at least about 0.90.

[0165] In some cases, the subject’s hearing classification may be determined and / or predicted with one or more machine learning algorithms and / or one or more predictive models with an Area Under the Receiver Operating Characteristic (AUROC) of up to about 0.65, up to about 0.70, up to about 0.75 up to about 0.80, up to about 0.82, up to about 0.84, up to about 0.86 up to about 0.88, or up to about 0.90.

[0166] An algorithm and / or predictive model can be implemented by way of software upon execution by the central processing unit 1605. In some cases, the predictive model may comprise a machine learning predictive model. In some cases, the machine learning predictive model may comprise one or more statistical, machine learning, and / or artificial intelligence algorithms. Examples of utilized algorithms, machine learning algorithms, and / or predictive models may include a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network (such as a deep neural network (DNN)), a recurrent neural network (RNN), a deep RNN, a long short-term memory (ESTM) recurrent neural network (RNN), decision tree algorithm, unsupervised clustering algorithm, a supervised clustering algorithm, unsupervised clustering algorithm, a regression algorithm, a gradient-boosting algorithm (e.g., a gradient-boosting implementation of a machine learning algorithm and / or predictive model such as a gradient-boosted decision trees), a gated recurrent unit (GRU), supervised learning algorithm, unsupervised learning algorithm, statistical, deep-learning algorithm for classification and regression, or any combination thereof. In some cases, the recurrent neural network may comprise units which can be LSTM units or GRU. In some cases, the predictive model and / or the machine learning algorithm may comprise an ensemble of one or more predictive models and / or machine learning algorithms.

[0167] The machine learning predictive model may likewise involve the estimation of ensemble models, comprised of multiple predictive models, and utilize techniques such as gradient boosting, for example in the construction of gradient-boosting decision trees. The machine learning predictive model may be trained using one or more training datasets corresponding to a subject’s hearing threshold audiogram, described elsewhere herein. In some embodiments, the one or more training datasets may comprise one or more subjects’ hearing threshold audiograms and corresponding audiogram classifications.

[0168] Training records may be constructed from sequences of observations. Such sequences may comprise a fixed length for ease of data processing. For example, sequences may be zero-padded or selected as independent subsets of a single subject’s records.

[0169] The one or more predictive models and / or one or more machine learning algorithms may process one or more input features to generate one or more output values comprising one or more subject’s hearing classification. For example, such hearing classifications may comprise a binary classification of a hearing within normal limits or hearing outside of normal limits (e.g., presence of an impaired hearing disorder), a classification between a group of categorical labels (e.g., ‘no hearing loss and / or hearing disorder’, ‘apparent hear loss or hearing disorder’, and ‘likely hearing loss or hearing disorder’), a likelihood (e.g., relative likelihood or probability) of developing hearing loss or hearing disorder, a score indicative of a presence of hearing loss or hearing disorder, or any combination thereof. Various predictive models and / or machine learning algorithms may be cascaded such that the output of one or more predictive models and / or one or more machine learning algorithms may be used as one or more input features to subsequent layers or subsections of the one or more predictive model and / or one or more machine learning algorithms.

[0170] In order to train the one or more predictive models and / or the one or more machine learning algorithms (e.g., by determining weights and correlations of the predictive model and / or the machine learning algorithm) to generate real-time classifications or predictions of one or more subjects’ hearing classifications, the model can be trained using datasets of subject’s hearing threshold audiograms (e.g., training datasets), described elsewhere herein. Such datasets may be sufficiently large to generate statistically significant classifications or predictions.

[0171] Datasets, as described elsewhere herein, may be split into subsets (e.g., discrete or overlapping), such as a training dataset, a development dataset, and a test dataset. For example, a dataset may be split into a training dataset comprising 80% of the dataset, a development dataset comprising 10% of the dataset, and a test dataset comprising 10% of the dataset. The training dataset may comprise about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80%, or about 90% of the dataset. The development dataset may comprise about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80%, or about 90% of the dataset. The test dataset may comprise about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80%, or about 90% of the dataset. Training sets (e.g., training datasets) may be selected by random sampling of a set of data corresponding to one or more subject hearing classification cohorts to ensure independence of sampling. In some cases, training sets (e.g., training datasets) may be selected by proportionate sampling of a set of data corresponding to one or more subject cohorts to ensure independence of sampling.

[0172] To improve the accuracy of predictive model and / or machine learning algorithm predictions and reduce overfitting of the predictive model and / or machine learning algorithm, the datasets may be augmented to increase the number of samples within the training set. For example, data augmentation may comprise rearranging the order of observations in a training record. To accommodate datasets having missing observations, methods to impute missing data may be used, such as forward-filling, back-filling, linear interpolation, and multi-task Gaussian processes. Datasets may be filtered to remove confounding factors. For example, within a database, a subset of subjects may be excluded.

[0173] Neural network techniques, such as dropout or regularization, may be used during training the one or more predictive models and / or one or more machine learning algorithms to prevent overfitting. The neural network may comprise a plurality of sub-networks, each of which is configured to generate a classification or prediction of a different type of output information (e.g., which may be combined to form an overall output of the neural network). The one or more predictive models and / or the one or more machine learning algorithms may alternatively utilize statistical and / or related algorithms comprising random forest, classification and regression trees, support vector machines, discriminant analyses, regression techniques, ensemble, gradient-boosted variations thereof, or any combination thereof.

[0174] When the one or more predictive models and / or the one or more machine learning algorithms generate a classification and / or a prediction of a subject’s hearing classification, a notification (e.g., alert or alarm) may be generated and transmitted to a health care provider, such as a physician, nurse, health care personnel managing, ear nose and throat physician, ear care practitioner, hearing specialist, or any combination thereof, treating a subject e.g., a subject within a hospital. Notifications may be transmitted via an automated phone call, a short message service (SMS), multimedia message service (MMS) message, an e-mail, an alert within a dashboard, or any combination thereof. The notification may comprise output information such as a prediction of a subject’s auditory health.

[0175] To validate the performance of the one or more predictive models and / or one more machine learning algorithms, different performance metrics may be generated. For example, an area under the receiver-operating curve (AUROC) may be used to determine the diagnostic and / or classification capability of the one or more predictive models and / or one or more machine learning algorithms. For example, the one or more predictive models and / or one or more machine learning algorithms may use classification thresholds which are adjustable, such that specificity and sensitivity are tunable, and the receiver-operating characteristic curve (ROC) can be used to identify the different operating points corresponding to different values of specificity and sensitivity of the one or more predictive models and / or one or more machine learning algorithms.

[0176] In some cases, such as when datasets are not sufficiently large, cross-validation may be performed to assess the robustness of one or more predictive models and / or one or more machine learning algorithms across different training and testing datasets.

[0177] To calculate performance metrics such as sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), AUPRC, AUROC, any combination thereof, or similar, the following definitions may be used. A “false positive” may refer to an outcome in which a positive outcome or result has been incorrectly or prematurely generated. A “true positive” may refer to an outcome in which positive outcome or result has been correctly generated. A “false negative” may refer to an outcome in which a negative outcome or result has been generated. A “true negative” may refer to an outcome in which a negative outcome or result has been generated.

[0178] The one or more predictive models and / or one or more machine learning algorithms may be trained until certain pre-determined conditions for accuracy or performance are satisfied, such as having minimum desired values corresponding to classification and / or diagnostic accuracy measures. For example, the diagnostic accuracy measure may correspond to prediction of a likelihood of occurrence of a hearing classification. Examples of diagnostic accuracy measures may include sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, area under the precision-recall curve (AUPRC), and area under the curve (AUC) of a Receiver Operating Characteristic (ROC) curve (AUROC) corresponding to the diagnostic accuracy of detecting or predicting a subject’s hearing classification.

[0179] For example, such a pre-determined condition may be that the sensitivity of predicting the subject’s hearing classification comprises a value of, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0180] As another example, such a pre-determined condition may be that the specificity of predicting the subject’s hearing classification comprises a value of, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0181] As another example, such a pre-determined condition may be that the positive predictive value (PPV) of predicting the subject’s hearing classification comprises a value of, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0182] As another example, such a pre-determined condition may be that the negative predictive value (NPV) of predicting the subject’s hearing classification comprises a value of, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0183] As another example, such a pre-determined condition may be that the area under the curve (AUC) of a Receiver Operating Characteristic (ROC) curve (AUROC) of predicting the subject’s hearing classification comprises a value of at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0184] As another example, such a pre-determined condition may be that the area under the precision-recall curve (AUPRC) of predicting the subject’s hearing classification comprises a value of at least about 0.10, at least about 0.15, at least about 0.20, at least about 0.25, at least about 0.30, at least about 0.35, at least about 0.40, at least about 0.45, at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0185] In some embodiments, the trained model may be trained or configured to predict the subject’s hearing classification with a sensitivity of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0186] In some embodiments, the trained model may be trained or configured to predict the subject’s hearing classification with a specificity of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0187] In some embodiments, the trained model may be trained or configured to predict the subject’s hearing classification with a positive predictive value (PPV) of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0188] In some embodiments, the trained model may be trained or configured to predict the subject’s hearing classification with a negative predictive value (NPV) of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0189] In some embodiments, the trained model may be trained or configured to predict the subject’s hearing classification with an area under the curve (AUC) of a Receiver Operating Characteristic (ROC) curve (AUROC) of at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0190] In some embodiments, the trained model may be trained or configured to predict the subject’s hearing classification with an area under the precision-recall curve (AUPRC) of at least about 0.10, at least about 0.15, at least about 0.20, at least about 0.25, at least about 0.30, at least about 0.35, at least about 0.40, at least about 0.45, at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0191] The training data sets may be collected from training subjects (e.g., humans). Each training subject has a diagnostic status indicating that they have either been diagnosed and / or classified with the subject’s hearing classification or have not been diagnosed with the subject’s hearing classification. The training procedure, as described elsewhere herein may be performed for each training subject in a plurality of training subjects.

[0192] In some embodiments, the machine learning analysis is performed by a device executing one or more programs (e.g., one or more programs stored in the Non-Persistent Memory or in the Persistent Memory) including instructions to perform the data analysis. In some embodiments, the data analysis is performed by a system comprising at least one processor (e.g., the processing core) and memory (e.g., one or more programs stored in the Non-Persistent Memory or in the Persistent Memory) comprising instructions to perform the data analysis.

[0193] Training the ML model may include, in some cases, selecting one or more untrained data models to train using a training data set. The selected untrained data models may comprise any type of untrained ML models for supervised, semi-supervised, self-supervised, and / or unsupervised machine learning. The selected untrained data models may be specified based upon input (e.g., user input) specifying relevant parameters to use as predicted variables and / or other variables to use as potential explanatory variables. For example, the selected untrained data models may be specified to generate an output (e.g., a prediction) based upon the input. Conditions for training the ML model from the selected untrained data models may likewise be selected, such as limits on the ML model complexity and / or limits on the ML model refinement past a certain point. The ML model may be trained (e.g., via a computer system such as a server) using the training data set. In some cases, a first subset of the training data set may be selected to train the ML model. The selected untrained data models may then be trained on the first subset of training data set using appropriate ML techniques, based upon the type of ML model selected and any conditions specified for training the ML model. In some cases, due to the processing power requirements of training the ML model, the selected untrained data models may be trained using additional computing resources (e.g., cloud computing resources). Such training may continue, in some cases, until at least one aspect of the ML model is validated and meets selection criteria to be used as a predictive model.

[0194] In some cases, one or more aspects of the ML model may be validated using a second subset of the training data set (e.g., distinct from the first subset of the training data set) to determine accuracy and robustness of the ML model. Such validation may include applying the ML model to the second subset of the training data set to make predictions derived from the second subset of the training data. The ML model may then be evaluated to determine whether performance is sufficient based upon the derived predictions. The sufficiency criteria applied to the ML model may vary depending upon the size of the training data set available for training, the performance of previous iterations of trained models, and / or user-specified performance requirements. If the ML model does not achieve sufficient performance, additional training may be performed. Additional training may include refinement of the ML model or retraining on a different first subset of the training dataset, after which the new ML model may again be validated and assessed. When the ML model has achieved sufficient performance, in some cases, the ML model may be stored for present and / or future use. The ML model may be stored as sets of parameter values and / or weights for analysis of further input (e.g., further relevant parameters to use as further predicted variables, further explanatory variables, further user interaction data, etc.), which may also include analysis logic and / or indications of model validity in some instances. In some cases, a plurality of ML models may be stored for generating predictions under different sets of input data conditions. In some embodiments, the ML model may be stored in a database (e.g., associated with a server).

[0195] In some embodiments, the one or more predictive models and / or the one or more machine learning algorithms may comprise a neural network or a convolutional neural network. See, Vincent et al., 2010, “Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion,” J Mach Learn Res 11, pp. 3371-3408; Larochelle etal., 2009, “Exploring strategies fortraining deep neural networks,” J Mach Learn Res 10, pp. 1-40; and Hassoun, 1995, Fundamentals of Artificial Neural Networks, Massachusetts Institute of Technology, each of which is hereby incorporated by reference.

[0196] SVMs are described in Cristianini and Shawe-Taylor, 2000, “An Introduction to Support Vector Machines,” Cambridge University Press, Cambridge; Boser etal., 1992, “A training algorithm for optimal margin classifiers,” in Proceedings of the 5th Annual ACM Workshop on Computational Learning Theory, ACM Press, Pittsburgh, Pa., pp. 142-152; Vapnik, 1998, Statistical Learning Theory, Wiley, New York; Mount, 2001, Bioinformatics: sequence and genome analysis, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.; Duda, Pattern Classification, Second Edition, 2001, John Wiley &Sons, Inc., pp. 259, 262-265; and Hastie, 2001, The Elements of Statistical Learning, Springer, New York; and Furey et al., 2000, Bioinformatics 16, 906-914, each of which is hereby incorporated by reference in its entirety. When used for classification, SVMs separate a given set of binary labeled data with a hyperplane that is maximally distant from the labeled data. For cases in which no linear separation is possible, SVMs can work in combination with the technique of'kernels', which automatically realizes a non-linear mapping to a feature space. The hyper-plane found by the SVM in feature space corresponds to a non-linear decision boundary in the input space.

[0197] Decision trees are described generally by Duda, 2001, Pattern Classification, John Wiley &Sons, Inc., New York, pp. 395-396, which is hereby incorporated by reference. Tree-based methods partition the feature space into a set of rectangles, and then fit a model (e.g., a constant) in each one. In some embodiments, the decision tree is random forest regression. One specific algorithm that can be used is a classification and regression tree (CART). Other specific decision tree algorithms comprise, ID3, C4.5, MART, and / or Random Forests. CART, ID3, and / or C4.5 are described in Duda, 2001, Pattern Classification, John Wiley &Sons, Inc., New York. pp. 396-408 and pp. 411-412, which is hereby incorporated by reference. CART, MART, and / or C4.5 are described in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York, Chapter 9, which is hereby incorporated by reference in its entirety. Random Forests are described in Breiman, 1999, “Random Forests—Random Features,” Technical Report 567, Statistics Department, U.C. Berkeley, September 1999, which is hereby incorporated by reference in its entirety.

[0198] Clustering (e.g., unsupervised clustering model algorithms and supervised clustering model algorithms) is described at pages 211-256 of Duda and Hart, Pattern Classification and Scene Analysis, 1973, John Wiley &Sons, Inc., New York, (hereinafter “Duda 1973”) which is hereby incorporated by reference in its entirety. As described in Section 6.7 of Duda 1973, the clustering problem may be described as one of finding natural groupings in a dataset. To identify natural groupings, two issues are addressed. First, a way to measure similarity (or dissimilarity) between two subjects is determined. This metric (similarity measure) may be used to ensure that the subjects in one cluster are more like one another than they are to subjects in other clusters. Second, a mechanism for partitioning the data into clusters using the similarity measure may be determined. Similarity measures are discussed in Section 6.7 of Duda 1973, where it is stated that one way to begin a clustering investigation may be to define a distance function and to compute the matrix of distances between all pairs of subjects in the training set. If distance is a good measure of similarity, then the distance between reference entities in the same cluster may be significantly less than the distance between the reference entities in different clusters. However, as stated on page 215 of Duda 1973, clustering may not require the use of a distance metric. For example, a nonmetric similarity function s(x, x') can be used to compare two vectors x and x'. Conventionally, s(x, x') is a symmetric function whose value may be large when x and x’ are somehow “similar.” An example of a nonmetric similarity function s(x, x’) is provided on page 218 of Duda 1973. Once a method for measuring “similarity” or “dissimilarity” between points in a dataset has been selected, clustering may require a criterion function that measures the clustering quality of any partition of the data. Partitions of the data set that extremize the criterion function may be used to cluster the data. See page 217 of Duda 1973. Criterion functions are discussed in Section 6.8 of Duda 1973. More recently, Duda et al.. Pattern Classification, 2nd edition, John Wiley &Sons, Inc. New York, has been published. Pages 537-563 describe clustering in detail. More information on clustering techniques can be found in Kaufman and Rousseeuw, 1990, Finding Groups in Data: An Introduction to Cluster Analysis, Wiley, New York, N.Y.; Everitt, 1993, Cluster analysis (3d ed.), Wiley, New York, N.Y.; and Backer, 1995, Computer-Assisted Reasoning in Cluster Analysis, Prentice Hall, Upper Saddle River, New Jersey, each of which is hereby incorporated by reference. Particular exemplary clustering techniques that can be used in the present disclosure include, but are not limited to, hierarchical clustering (e.g., agglomerative clustering using nearest-neighbor algorithm, farthest-neighbor algorithm, the average linkage algorithm, the centroid algorithm, and / or the sum-of-squares algorithm), k-means clustering, fuzzy k-means clustering algorithm, Jarvis-Patrick clustering, or any combination thereof. In some embodiments, the clustering comprises unsupervised clustering, where no preconceived notion of what clusters may form when the training set is clustered, are imposed.

[0199] Regression models, such as that of the multi-category logit models, are described in Agresti, An Introduction to Categorical Data Analysis, 1996, John Wiley &Sons, Inc., New York, Chapter 8, which is hereby incorporated by reference in its entirety. In some embodiments, the one or more predictive model and / or one or more machine learning algorithms may make use of a regression model disclosed in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York, which is hereby incorporated by reference in its entirety. In some embodiments, gradient-boosting models may be used toward, for example, the classification algorithms described herein; these gradient-boosting models are described in Boehmke, Bradley; Greenwell, Brandon (2019). "Gradient Boosting". Hands-On Machine Learning with R Chapman Hall. pp. 221-245. ISBN978-1-138-49568-5., which is hereby incorporated by reference in its entirety. In some embodiments, ensemble modeling techniques may be used, for example, toward the classification algorithms described herein; these ensemble modeling techniques are described in the implementation of classification models herein, are described in Zhou Zhihua (2012). Ensemble Methods: Foundations and Algorithms. Chapman andHall / CRC. ISBN 978-1-439-83003-1, which is hereby incorporated by reference in its entirety.

[0200] While preferred embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. It is not intended that the invention be limited by the specific examples provided within the specification. While the invention has been described with reference to the aforementioned specification, the descriptions and illustrations of the embodiments herein are not meant to be construed in a limiting sense. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the invention. Furthermore, it shall be understood that all aspects of the invention are not limited to the specific depictions, configurations or relative proportions set forth herein which depend upon a variety of conditions and variables. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention. It is therefore contemplated that the invention shall also cover any such alternatives, modifications, variations, or equivalents. It is intended that the following claims define the scope of the invention and that methods and structures within the scope of these claims and their equivalents be covered thereby.

[0201] Although the above steps show each of the methods or sets of operations in accordance with embodiments, a person of ordinary skill in the art will recognize many variations based on the teaching described herein. The steps may be completed in a different order. Steps may be added or omitted. Some of the steps may comprise sub-steps. Many of the steps may be repeated as often as beneficial. One or more of the steps of each of the methods or sets of operations may be performed with circuitry as described herein, for example, one or more of the processor or logic circuitry such as programmable array logic for a field programmable gate array. The circuitry may be programmed to provide one or more of the steps of each of the methods or sets of operations, and the program may comprise program instructions stored on a computer readable memory or programmed steps of the logic circuitry such as the programmable array logic or the field programmable gate array, for example.

[0202] Whenever the term “at least,” “greater than,” or “greater than or equal to” precedes the first numerical value in a series of two or more numerical values, the term “at least,” “greater than” or “greater than or equal to” applies to each of the numerical values in that series of numerical values. For example, greater than or equal to 1, 2, or 3 is equivalent to greater than or equal to 1, greater than or equal to 2, or greater than or equal to 3.

[0203] Whenever the term “no more than,” “less than,” or “less than or equal to” precedes the first numerical value in a series of two or more numerical values, the term “no more than,” “less than,” or “less than or equal to” applies to each of the numerical values in that series of numerical values. For example, less than or equal to 3, 2, or 1 is equivalent to less than or equal to 3, less than or equal to 2, or less than or equal to 1.

[0204] Certain inventive embodiments herein contemplate numerical ranges. When ranges are present, the ranges include the range endpoints. Additionally, every sub range and value within the range is present as if explicitly written out. The term “about” or “approximately” may mean within an acceptable error range for the particular value, which will depend in part on how the value is measured or determined, e.g., the limitations of the measurement system. For example, “about” may mean within 1 or more than 1 standard deviation, per the practice in the art. Alternatively, “about” may mean a range of up to 20%, up to 10%, up to 5%, or up to 1% of a given value. Where particular values are described in the application and claims, unless otherwise stated the term “about” meaning within an acceptable error range for the particular value may be assumed.

[0205] The disclosure of this application also contains the following numbered clauses: 1. A method of providing an auditory health recommendation to a subject, comprising one or more of: (i) providing one or more audio stimuli to a subject’s ear with a computing device; (ii) receiving one or more responses from said subject in response to said one or more audio stimuli with said computing device; (iii) determining a hearing classification of said subject from said one or more responses; and (iv) providing said auditory health recommendation to said subject based on said hearing classification. 2. The method of clause 1, wherein said auditory health recommendation comprises a referral to a health care professional. 3. The method of clauses 1 or 2, wherein said auditory health recommendation is provided to said subject on the computing device, a different personal computing device, website portal, personal computer, or any combination thereof. 4. The method of any one of clauses 1-3, wherein said one or more audio stimuli are provided to said subject’s ear through a headphone or ear bud. 5. The method of clause 4, wherein said headphone or said ear bud are wired or wireless. 6. The method of any one of clauses 1-5, wherein said determining said hearing classification of said subject comprises providing said one or more responses of said subject as an input to one or more trained machine learning algorithms or one or more predictive models that output said hearing classification of said subject. 7. The method of clause 6, wherein said one or more trained machine learning algorithms or said one or more predictive models are trained with one or more responses of a plurality of subjects in response to one or more audio stimuli and a corresponding hearing classification of said plurality of subjects. 8. The method of any one of clauses 1-7, wherein said one or more audio stimuli comprise audio stimuli of one or more frequencies and / or one or more intensities. 9. The method of clause 8, wherein an audio stimulus of said one or more audio stimuli is provided to said subject at a frequency of said one or more frequencies with at least 20 occurrences, and wherein each occurrence of said at least 20 occurrences comprises a different intensity of said audio stimulus. 10. The method of clauses 8 or 9, wherein said one or more frequencies comprise up to about 4 frequencies or up to about 8 frequencies. 11. The method of any one of clauses 8-10, wherein said one or more frequencies comprise: about 250 Hz, about 500 Hz, about 1000 Hz, about 2000 Hz, about 3000 Hz, about 4000 Hz, about 6000Hz, about 8000Hz, or any combination thereof. 12. The method of any one of clauses 8-11, wherein said one or more frequencies comprise one or more frequency bands. 13. The method of clause 12, wherein said one or more frequency bands comprise frequency ranges of about 100Hz to about 999Hz, about 1000Hz to about 3999Hz, about 4000Hz to about 8000Hz, or any combination thereof. 14. The method of any one of clauses 8-13, wherein said subject’s hearing classification is determined from said subject’s one or more responses to at least 4 frequencies of said one or more frequencies of said one or more audio stimuli. 15. The method of any one of clauses 8-14, wherein said subject’s hearing classification is determined from an average of said subject’s one or more responses from at about least 2 frequency bands, at least about 3 frequency bands, or at least 4 frequency bands of said one or more frequencies of said one or more audio stimuli. 16. The method of any one of clauses 1-15, wherein said one or more responses from said subject comprise an intensity of said one or more audio stimuli at which said subject provided a response of detecting said one or more audio stimuli. 17. The method of clause 16, wherein said subject’s hearing classification comprises normal hearing when said intensity comprises up to about 20dB, up to about 25dB, or up to about 40dB. 18. The method of clauses 16 or 17, wherein said hearing classification of said subject comprises normal hearing when said intensity of said one or more audio stimuli comprises about -lOdB to about 25 dB. 19. The method of any one of clauses 16-18, wherein said hearing classification of said subject comprises mild hearing loss when said intensity of said one or more audio stimuli comprises about 26dB to about 39dB. 20. The method of any one of clauses 16-19, wherein said hearing classification of said subject comprises moderate hearing loss when said intensity of said one or more audio stimuli comprises about 40dB to about 69dB. 21. The method of any one of clauses 16-20, wherein said hearing classification of said subject comprises severe hearing loss when said intensity of said one or more audio stimuli comprises about 70dB to about 94dB. 22. The method of any one of clauses 16-21, wherein said hearing classification of said subject comprises profound hearing loss when said intensity of said one or more audio stimuli comprises about 95dB to about 120dB. 23. The method of any one of clauses 1-22, wherein said determining said hearing classification comprises determining whether said subject has one or more of: normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, and / or an unfamiliar hearing profile. 24. The method of any one of clauses 1-23, wherein said subject’s ear comprises a first ear and a second ear, wherein said one or more responses comprise a first set of one or more responses from said one or more audio stimuli provided to said first ear and a second set of one or more responses from providing said one or more audio stimuli to said second ear. 25. The method of clause 24, wherein said subject’s hearing classification comprises asymmetric hearing when, for each of two or more frequencies, intensity thresholds of the responses at the respective frequency differ by at least about 15 dB between the first and second ears of the subject. 26. The method of any one of clauses 1-25, further comprising displaying an object overlaid on a graph of said one or more responses of said subject when a reference audiogram is selected. 27. The method of any one of clauses 1-26, wherein determining said hearing classification of said subject comprises comparing said one or more responses of said subject to a library of one or more responses to said one or more audio stimuli associated with one or more hearing classifications. 28. The method of any one of clauses 1-27, further comprising providing a period of time where no audio stimulation is provided to said subject’s ear. 29. The method of clause 28, wherein said period of time comprises at least about 2 seconds or at least about 3 seconds. 30. The method of any one of clauses 1-29, wherein said recommendation of said subject is uploaded to a server or cloud base storage. 31. The method of any one of clauses 1-30, further comprising removing ear wax from said subject’s ear. 32. The method of clause 31, wherein said removing of said ear wax is conducted before providing said one or more audio stimuli to said subject’s ear. 33. The method of any one of clauses 1-32, wherein said one or more audio stimuli comprise pure tones. 34. The method of any one of clauses 1-33, wherein said one or more responses of said subject comprise pressing a surface when said subject detects or hears said one or more audio stimuli. 35. The method of clause 34, wherein said surface comprises a surface of an interface on a computing device. 36. The method of clause 35, wherein said computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. 37. The method of any one of clauses 1-36, further comprising generating or updating an auditory health profde of said subject based on one or more of said determined hearing classification or said auditory health recommendation provided based thereon. 38. The method of any one of clauses 1-37, repeating steps (i) to (iv) at one or more regular intervals. 39. A system for providing an auditory health recommendation to a subject, comprising: one or more processors and memory storing one or more programs for execution by said one or more processors, said one or more programs comprising instructions to perform the method of any one of clauses 1-38. 40. A system for providing an auditory health recommendation to a subject, comprising: one or more processors and memory storing one or more programs for execution by said one or more processors, said one or more programs comprising instructions to: (i) provide one or more audio stimuli to said subject’s ear with a computing device; (ii) receive one or more responses from said subject in response to said one or more audio stimuli with said computing device; (iii) determine a hearing classification of said subject from said one or more responses with said one or more processors; and (iv) provide said auditory health recommendation to said subject based on said hearing classification. 41. The system of clause 40, wherein said recommendation comprises a referral to a health care professional. 42. The system of clauses 40 or 41, wherein said recommendation is provided to said subject on said computing device, a different personal device, website portal, personal computer, or any combination thereof. 43. The system of any one of clauses 40-42, wherein said one or more audio stimuli are provided to said subject’s ear through a headphone or ear bud. 44. The system of clause 43, wherein said headphone or said ear bud are wired or wireless. 45. The system of any one of clauses 40-44, wherein said determining said hearing classification of said subject comprises providing said one or more responses of said subject as an input to one or more trained machine learning algorithms or one or more predictive models that output said hearing classification of said subject. 46. The system of clause 45, wherein said one or more trained machine learning algorithms or said one or more predictive models are trained with one or more responses of a plurality of subjects in response to one or more audio stimuli and a corresponding hearing classification of said plurality of subjects. 47. The system of any one of clauses 40-46, wherein said one or more audio stimuli comprise audio stimuli of one or more frequencies and / or one or more intensities. 48. The system of clause 47, wherein an audio stimulus of said one or more audio stimuli is provided to said subject at a frequency of said one or more frequencies with at least 20 occurrences, and wherein each occurrence of said at least 20 occurrences comprises a different intensity of said audio stimulus. 49. The system of clauses 47 or 48, wherein said one or more frequencies comprise up to about 4 frequencies or up to about 8 frequencies. 50. The system of any one of clauses 47-49, wherein said one or more frequencies comprise: about 125Hz, about 250Hz, about 400Hz, about 750Hz, about 1000Hz, about 1500Hz, about 2000Hz, about 4000Hz, about 6000Hz, about 8000Hz, about 10,000Hz, about 12,500Hz, about 16,000Hz, or any combination thereof. 51. The system of any one of clauses 47-50, wherein said one or more frequencies comprise one or more frequency bands. 52. The system of any one of clauses 47-51, wherein said one or more frequency bands comprise frequency ranges of about 100Hz to about 999Hz, about 1000Hz to about 3999Hz, about 4000Hz to about 8000Hz, or any combination thereof. 53. The system of any one of clauses 47-52, wherein said subject’s hearing classification is determined from said subject’s one or more responses to at least 4 frequencies of said one or more frequencies of said one or more audio stimuli. 54. The system of any one of clauses 47-53, wherein said subject’s hearing classification is determined from an average of said subject’s one or more responses from at about least 2 frequency bands, at least about 3 frequency bands, or at least 4 frequency bands of said one or more frequencies of said one or more audio stimuli. 55. The system of any one of clauses 40-54, wherein said one or more responses from said subject comprise an intensity of said one or more audio stimuli at which said subject provided a response of detecting said one or more audio stimuli. 56. The system of clause 55, wherein said subject’s hearing classification comprises normal hearing when said intensity comprises up to about 20dB, up to about 25dB, or up to about 40dB. 57. The system of clauses 55 or 56, wherein said hearing classification of said subject comprises normal hearing when said intensity of said one or more audio stimuli comprises about -lOdB to about 25 dB. 58. The system of any one of clauses 55-57, wherein said hearing classification of said subject comprises mild hearing loss when said intensity of said one or more audio stimuli comprises about 26dB to about 39dB. 59. The system of any one of clauses 55-58, wherein said hearing classification of said subject comprises moderate hearing loss when said intensity of said one or more audio stimuli comprises about 40dB to about 69dB. 60. The system of any one of clauses 55-59, wherein said hearing classification of said subject comprises severe hearing loss when said intensity of said one or more audio stimuli comprises about 70dB to about 94dB. 61. The system of any one of clauses 55-60, wherein said hearing classification of said subject comprises profound hearing loss when said intensity of said one or more audio stimuli comprises about 95dB to about 120dB. 62. The system of any one of clauses 40-61, wherein said determining said hearing classification comprises determining whether said subject has normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, or an unfamiliar hearing profile. 63. The system of any one of clauses 40-62, wherein said subject’s ear comprises a first ear and a second ear, wherein said one or more responses comprise a first set of one or more responses from said one or more audio stimuli provided to said first ear and a second set of one or more responses from providing said one or more audio stimuli to said second ear. 64. The system of clause 63, wherein said subject’s hearing classification comprises asymmetric hearing when, for each of two or more frequencies, intensity thresholds of the responses at the respective frequency differ by at least about 15 dB between the first and second ears of the subject. 65. The system of any one of clauses 40-64, further comprising displaying an object overlaid on a graph of said one or more responses of said subject when a reference audiogram is selected. 66. The system of any one of clauses 40-65, wherein determining said hearing classification of said subject comprises comparing said one or more responses of said subject to a library of one or more responses to said one or more audio stimuli associated with one or more hearing classifications. 67. The system of any one of clauses 40-66, further comprising providing a period of time where no audio stimulation is provided to said subject’s ear. 68. The system of clause 67, wherein said period of time comprises at least about 2 seconds or at least about 3 seconds. 69. The system of any one of clauses 40-68, wherein said recommendation of said subject is uploaded to a server or cloud base storage. 70. The system of any one of clauses 40-69, further comprising removing ear wax from said subject’s ear. 71. The system of clause 70, wherein said removing of said ear wax is conducted before providing said one or more audio stimuli to said subject’s ear. 72. The system of any one of clauses 40-71, wherein said one or more audio stimuli comprise pure tones. 73. The system of any one of clauses 40-72, wherein said one or more responses of said subject comprise pressing a surface when said subject detects or hears said one or more audio stimuli. 74. The system of any one of clauses 40-73, wherein said surface comprises a surface of a user interface on said computing device. 75. The system of any one of clauses 40-74, wherein said computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. 76. The system of any one of clauses 40-75, wherein said one or more programs further comprises instructions to generate or update an auditory health profile of said subject based on one or more of said determined hearing classification or said auditory health recommendation provided based thereon. 77. The system of any one of clauses 39-76, repeating steps (i) to (iv) at one or more regular intervals. 78. A method for providing a subject’s hearing referral to a health care provider, comprising: providing said subject’s hearing referral to said health care provider from a computing system in response to said computer system receiving said subject’s hearing classification and indication for referral, wherein said subject’s hearing classification is determined by: providing one or more audio stimuli to said subject’s ear with a computing device; receiving one or more responses from said subject in response to said one or more audio stimuli with said computing device; and determining said hearing classification of said subject from said one or more responses. 79. The method of clause 78, wherein said computing system comprises a smart phone, a server, a web portal, a personal computer, a laptop computer, a tablet, or any combination thereof. 80. The method of clause 79, wherein said computing system is coupled to a headphone or ear bud. 81. The method of clauses 78 or 79, wherein said one or more responses from said subject comprise an intensity of said one or more audio stimuli at which said subject provided a response of detecting or hearing said one or more audio stimuli. 82. The method of any one of clauses 78-81, wherein said determining said hearing classification comprises determining whether said subject has one or more of: normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, and / or an unfamiliar hearing profile. 83. The method of any one of clauses 78-82, wherein said subject’s hearing classification is further determined by removing said subject’s ear wax before providing said one or more audio stimuli to said subject’s ear. 84. The method of any one of clauses 78-82, wherein said one or more responses of said subject comprise pressing a surface when said subject detects or hears said one or more audio stimuli. 85. The method of clause 84, wherein said surface comprises a surface of a user interface on said computing device. 86. The method of any one of clauses 78-85, wherein said computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. 87. The method of any one of clauses 78-86, wherein said determining said hearing classification of said subject comprises providing said one or more responses of said subject as an input to one or more trained machine learning algorithms or one or more predictive models that output said hearing classification of said subject. 88. The method of any one of clauses 78-87, wherein said hearing referral comprises clinical data of said subject, contact information of said subject, said hearing classification, or any combination thereof. 89. The method of any one of clauses 78-88, further comprising accepting said subject’s hearing referral. 90. A system for providing an auditory health recommendation to a subject, comprising: one or more processors and memory storing one or more programs for execution by said one or more processors, said one or more programs comprising instructions to perform the method of any one of clauses 78-89. 91. A system for providing a subject’s hearing referral to a health care provider, comprising: one or more processors and memory storing one or more programs for execution by said one or more processors, said one or more programs comprising instructions to: provide said subject’s hearing referral to said health care provider from a computing system in response to said computing system receiving said subject’s hearing classification and indication for referral from a computing device, wherein said subject’s hearing classification is determined by: providing one or more audio stimuli to said subject’s ear with a computing device; receiving one or more responses from said subject in response to said one or more audio stimuli with said computing device; and determining said hearing classification of said subject from said one or more responses. 92. The system of clause 91, wherein said computing system comprises a smart phone, a server, a web portal, a personal computer, a laptop computer, a tablet, or any combination thereof. 93. The system of clause 92, wherein said computing system is coupled to a headphone or ear bud. 94. The system of clauses 91 or 92, wherein said one or more responses from said subject comprise an intensity of said one or more audio stimuli at which said subject provided a response of detecting or hearing said one or more audio stimuli. 95. The system of any one of clauses 91-94, wherein said determining said hearing classification comprises determining whether said subject has one or more of: normal hearing, normal hearing, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, and / or an unfamiliar hearing profile. 96. The system of any one of clauses 91-95, wherein said subject’s hearing classification is further determined by removing said subject’s ear wax before providing said one or more audio stimuli to said subject’s ear. 97. The system of any one of clauses 91-96, wherein said one or more responses of said subject comprise pressing a surface when said subject detects or hears said one or more audio stimuli. 98. The system of clause 97, wherein said surface comprises a surface of a user interface on said computing device. 99. The system of any one of clauses 91-98, wherein said computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. 100. The system of any one of clauses 91-99, wherein said determining said hearing classification of said subject comprises providing said one or more responses of said subject as an input to one or more trained machine learning algorithms or one or more predictive models that output said hearing classification of said subject. 101. The system of any one of clauses 91-100, wherein said hearing referral comprises clinical data of said subject, contact information of said subject, said hearing classification, or any combination thereof. 102. The system of any one of clauses 91-100, wherein said instructions further comprise accepting said subject’s hearing referral. 103. A method fortraining a machine learning algorithm or predictive model, comprising: receiving or obtaining one or more training subject responses to one or more audio stimuli provided to one or more training subjects’ ear(s) and one or more corresponding hearing classifications of said one or more training subjects; and training said machine learning algorithm or predictive model with said one or more responses and corresponding hearing classifications thereby generating a trained machine learning algorithm or trained predictive model. 104. The method of clause 103, wherein said one or more training subject responses to said one or more audio stimuli comprise an intensity of said one or more audio stimuli that said one or more training subjects detect or hear said one or more audio stimuli. 105. The method of clauses 103 or 104, wherein said machine learning algorithm or predictive model comprises a neural network, a support vector machine, a random forest model, a naive Bayes classification algorithm, a gradient-boosting algorithm, linear regression algorithm, unsupervised clustering algorithm, supervised clustering algorithm, or any combination thereof. 106. The method of any one of clauses 103-105, wherein said trained machine learning algorithm or said trained predictive model is configured to predict or determine one or more subjects’ hearing classification with at least about 85% accuracy, specificity, sensitivity, or any combination thereof. 107. The method of any one of clauses 103-106, wherein said one or more training subject responses comprise one or more responses of said one or more training subjects to said one or more audio stimuli provided to a left ear and / or a right ear of said one or more training subjects’ ears. 108.The method of any one of clauses 103-107, wherein said one or more audio stimuli are provided to said one or more training subjects by a computing device. 109. The method of clause 108, wherein said computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. 110. The method of clauses 108-109, wherein said one or more training subject responses comprise pressing a surface when said one or more training subjects detect or hear said one or more audio stimuli. 111 .The method of clause 110, wherein said surface comprises a surface of said computing device. 112. The method of any one of clauses 103-109, wherein said trained machine learning algorithm or said trained predictive model is configured to receive an input of a subject’s one or more responses of detecting or hearing one or more audio stimuli provided to said subject’s ear, and wherein said trained machine learning algorithm or said trained predictive model is configured to output a hearing classification of said subject. 113. The method of any one of clauses 103-112, wherein said trained machine learning algorithm or said trained predictive model comprises a plurality of trained machine learning algorithms or a plurality of trained predictive models. 114. A system for providing an auditory health recommendation to a subject, comprising: one or more processors and memory storing one or more programs for execution by said one or more processors, said one or more programs comprising instructions to perform the method of anyone of clauses 103-113. 115. A system configured to train a machine learning algorithm or predictive model, comprising: one or more processors and memory storing one or more programs for execution by said one or more processors, said one or more programs comprising instructions to: receive or obtain one or more training subject responses to one or more audio stimuli provided to said one or more training subjects’ ear(s) and corresponding one or more hearing classification of said one or more training subjects; and train said machine learning algorithm or predictive model with said one or more responses and corresponding hearing classifications thereby generating a trained machine learning algorithm or trained predictive model. 116. The system of clause 115, wherein said one or more training subject responses to one or more audio stimuli comprise an intensity of said one or more audio stimuli that said one or more training subjects detect or hear said one or more audio stimuli. 117. The system of clauses 115 or 116, wherein said machine learning algorithm or predictive model comprises a neural network, support vector machine, random forest, naive Bayes classification algorithm, gradient-boosting algorithm, linear regression algorithm, unsupervised clustering algorithm, supervised clustering algorithm, or any combination thereof. 118.The system of any one of clauses 115-117, wherein said trained machine learning algorithm or said trained predictive model is configured to predict or determine one or more subjects’ hearing classification with at least about 85% accuracy, specificity, sensitivity, or any combination thereof. 119. The system of any one of clauses 115-118, wherein said one or more training subject responses comprise one or more responses of said one or more training subjects to said one or more audio stimuli provided to a left ear and / or a right ear of said one or more training subjects’ ears. 120. The system of any one of clauses 115-119, wherein said one or more audio stimuli are provided to said one or more training subjects by a computing device. 121. The system of clause 120, wherein said computing device comprises a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof. 122. The system of clauses 120 or 121, wherein said one or more training subject responses comprise pressing a surface when said one or more training subjects detect or hear said one or more audio stimuli. 123.The system of clause 122, wherein said surface comprises a surface of said computing device. 124. The system of any one of clauses 115-123, wherein said trained machine learning algorithm or said trained predictive model is configured to receive an input of a subject’s one or more response of detecting or hearing one or more audio stimuli provided to said subject, and wherein said trained machine learning algorithm or said trained predictive model is configured to output a hearing classification of said subject. 125. The system of any one of clauses 115-124, wherein said trained machine learning algorithm or said trained predictive model comprises a plurality of trained machine learning algorithms or a plurality of trained predictive models. 126. A method of providing an auditory health recommendation to a subject, comprising: (i) receiving one or more responses of a subject to one or more audio stimuli provided to the subject via a computing device; (ii) determining a hearing classification of the subject based on the one or more responses; and (iii) outputting an auditory health recommendation to the subject based on the hearing classification. 127. The method of clause 126, further comprising the steps of any one of clauses 1 to 38. 128.The method of clause 126 or 127, wherein the one or more audio stimuli comprise audio stimuli of one or more frequencies and / or one or more intensities, wherein: an audio stimulus of said one or more audio stimuli is provided to said subject at a frequency of said one or more frequencies with at least 20 occurrences, and wherein each occurrence of said at least 20 occurrences comprises a different intensity of said audio stimulus, optionally wherein said one or more frequencies comprise up to about 4 frequencies or up to about 8 frequencies; said one or more frequencies comprise: about 250 Hz, about 500 Hz, about 1000 Hz, about 2000 Hz, about 3000 Hz, about 4000 Hz, about 6000Hz, about 8000Hz, or any combination thereof; and / or said one or more frequencies comprise one or more frequency bands, optionally wherein said one or more frequency bands comprise frequency ranges of about 100Hz to about 999Hz, about 1000Hz to about 3999Hz, about 4000Hz to about 8000Hz, or any combination thereof. 129. A method for providing a subject’s hearing referral to a health care provider, comprising: providing said subject’s hearing referral to said health care provider from a computing system in response to said computer system receiving said subject’s hearing classification and indication for referral, wherein said subject’s hearing classification is determined by: receiving one or more responses of a subject to one or more audio stimuli provided to the subject via a computing device; and determining a hearing classification of the subject based on the one or more responses. 130. The method of clauses 6 or 87 or the system of clauses 45 or 100, or any clause dependent thereon, wherein the one or more trained machine learning algorithms or one or more predictive models are trained according to the method of any of clauses 103 to 113. 131.Apparatus comprising: means for providing one or more audio stimuli to a subject’s ear with a computing device; means for receiving one or more responses from said subject in response to said one or more audio stimuli with said computing device; means for determining a hearing classification of said subject from said one or more responses; and means for providing said auditory health recommendation to said subject based on said hearing classification. 132.Apparatus comprising: means for receiving one or more responses of a subject to one or more audio stimuli provided to the subject via a computing device; means for determining a hearing classification of the subject based on the one or more responses; and means for providing said auditory health recommendation to the subject based on the hearing classification. 133.Apparatus comprising: means for providing a subject’s hearing referral to a health care provider from a computing system in response to said computer system receiving said subject’s hearing classification and indication for referral, wherein said subject’s hearing classification is determined by: providing one or more audio stimuli to said subject’s ear with a computing device; receiving one or more responses from said subject in response to said one or more audio stimuli with said computing device; and determining said hearing classification of said subject from said one or more responses. 134.Apparatus comprising: means for providing a subject’s hearing referral to a health care provider from a computing system in response to said computer system receiving said subject’s hearing classification and indication for referral, wherein said subject’s hearing classification is determined by: receiving one or more responses of a subject to one or more audio stimuli provided to the subject via a computing device; and determining a hearing classification of the subject based on the one or more responses. 5 13 5. Apparatus comprising: means for receiving or obtaining one or more training subject responses to one or more audio stimuli provided to one or more training subjects’ ear(s) and one or more corresponding hearing classifications of said one or more training subjects; and means for training said machine learning algorithm or predictive model with said one or more 10 responses and corresponding hearing classifications thereby generating a trained machine learning algorithm or trained predictive model. 136. A computer program, computer program product or computer readable medium comprising software code adapted, when executed by a computer system, to perform a method as set out in any of clauses 1-38, 78-89, 103-113 or 126-130.

Claims

1. A method of providing an auditory health recommendation to a subject, comprising:(i) providing one or more audio stimuli to a subject’s ear with a computing device;(ii) receiving one or more responses from said subject in response to said one or more audio stimuli with said computing device;(iii) determining a hearing classification of said subject from said one or more responses; and (iv)providing said auditory health recommendation to said subject based on said hearing classification.

2. The method of claim 1, wherein said auditory health recommendation comprises a referral to a health care professional, and / or wherein said auditory health recommendation is provided to said subject on the computing device, a different personal computing device, website portal, personal computer, or any combination thereof.

3. The method of claim 1 or 2, wherein said one or more audio stimuli are provided to said subject’s ear through a headphone or ear bud; optionally wherein said headphone or said ear bud are wired or wireless.

4. The method of any preceding claim, wherein said determining said hearing classification of said subject comprises providing said one or more responses of said subject as an input to one or more trained machine learning algorithms or one or more predictive models that output said hearing classification of said subject.

5. The method of claim 4, wherein said one or more trained machine learning algorithms or said one or more predictive models are trained with one or more responses of a plurality of subjects in response to one or more audio stimuli and a corresponding hearing classification of said plurality of subjects.

6. The method of any preceding claim, wherein said one or more audio stimuli comprise audio stimuli of one or more frequencies and / or one or more intensities.

7. The method of claim 6. wherein an audio stimulus of said one or more audio stimuli is provided to said subject at a frequency of said one or more frequencies with at least 20 occurrences, and wherein each occurrence of said at least 20 occurrences comprises a different intensity of said audio stimulus; and / or wherein said one or more frequencies comprise up to about 4 frequencies or up to about 8 frequencies.

8. The method of claim 6 or 7, wherein said one or more frequencies comprise one or more frequency bands.

9. The method of any one of claims 6-8, wherein said subject’s hearing classification is determined from said subject’s one or more responses to at least 4 frequencies of said one or more frequencies of said one or more audio stimuli.

10. The method of any one of claims 6-9, wherein said subject’s hearing classification is determined from an average of said subject’s one or more responses from at about least 2 frequency bands, at least about 3 frequency bands, or at least 4 frequency bands of said one or more frequencies of said one or more audio stimuli.

11. The method of any preceding claim, wherein said one or more responses from said subject comprise an intensity of said one or more audio stimuli at which said subject provided a response of detecting said one or more audio stimuli; optionally wherein said subject’s hearing classification comprises normal hearing when said intensity comprises up to about 20dB, up to about 25dB, or up to about 40dB.

12. The method of claim 11, wherein said hearing classification of said subject comprises: normal hearing when said intensity of said one or more audio stimuli comprises about -lOdB to about 25dB;mild hearing loss when said intensity of said one or more audio stimuli comprises about 26dB to about 39dB;moderate hearing loss when said intensity of said one or more audio stimuli comprises about 40dB to about 69dB;severe hearing loss when said intensity of said one or more audio stimuli comprises about 70dB to about 94dB; and / orprofound hearing loss when said intensity of said one or more audio stimuli comprises about 95dB to about 120dB.

13. The method of any preceding claim, wherein said determining said hearing classification comprises determining whether said subject has normal hearing, hearing loss, mild hearing loss, moderate hearing loss, severe hearing loss, asymmetric hearing, an inconclusive result, or an unfamiliar hearing profile.

14. The method of any preceding claim, wherein said subject’s ear comprises a first ear and a second ear, wherein said one or more responses comprise a first set of one or more responses from said one or more audio stimuli provided to said first ear and a second set of one or more responses from providing said one or more audio stimuli to said second ear.

15. The method of claim 14, wherein said subject’s hearing classification comprises asymmetric hearing when, for each of two or more frequencies, intensity thresholds of the responses at the respective frequency differ by at least about 15 dB between the first and second ears of the subject.

16. The method of any preceding claim, further comprising displaying an object overlaid on a graph of said one or more responses of said subject when a reference audiogram is selected.

17. The method of any preceding claim, wherein determining said hearing classification of said subject comprises comparing said one or more responses of said subject to a library of one or more responses to said one or more audio stimuli associated with one or more hearing classifications.

18. The method of any preceding claim, further comprising providing a period of time where no audio stimulation is provided to said subject’s ear; optionally wherein said period of time comprises at least about 2 seconds or at least about 3 seconds.

19. The method of any preceding claim, wherein said recommendation of said subject is uploaded to a server or cloud base storage.

20. The method of any preceding claim, further comprising removing ear wax from said subject’s ear; optionally wherein said removing of said ear wax is conducted before providing said one or more audio stimuli to said subject’s ear.

21. The method of any preceding claim, wherein said one or more audio stimuli comprise pure tones.

22. The method of any preceding claim, wherein said one or more responses of said subject comprise pressing a surface when said subject detects or hears said one or more audio stimuli; optionally wherein said surface comprises a surface of an interface on a computing device such as a smartphone, tablet, laptop computer, personal computer, web interface, or any combination thereof.

23. The method of any preceding claim, further comprising generating or updating an auditory health profde of said subject based on one or more of said determined hearing classification or said auditory health recommendation provided based thereon.

24. The method of any preceding claim, repeating steps (i) to (iv) at one or more regular intervals.

25. A system for providing an auditory health recommendation to a subject, comprising: one or more processors and memory storing one or more programs for execution by said one or more processors, said one or more programs comprising instructions to perform the method of any preceding claim.

Images