Search Results

The Hearing Healthcare Pro software will capture audiometry test results, and will allow user input of tympanometry results, physical examination results, and patient history. The software will summarize the inputted data and provide a "normal" or "abnormal" determination. If the decision is "abnormal", the software can make a recommendation for clinical follow-up. The dB Diagnostic Systems, Inc. software is for adult use only and may be used by healthcare providers not normally trained or experienced in hearing, audiology, or otology, such as primary care providers. The Hearing Healthcare Pro software is not intended to make a clinical diagnosis.

The Hearing Healthcare (HHC) Pro is software designed to be used by healthcare providers (HCP) either trained and experienced in hearing, audiology, or otology, such as Ear, Nose, and Throat (ENT) doctors or not, such as primary care providers. The HHC Pro software is not intended to make a clinical diagnosis. The HHC Pro software is a screening product. The software aggregates the inputted data and determines whether the person's hearing is considered normal or abnormal. If the results are abnormal, the software can either provide a recommendation for further clinical consultation with a licensed physician with expertise in Ear, Nose and Throat (ENT), or allow for an ENT-trained physician to enter their own recommendation in a blank text field. Further, the software can distinguish between symmetrical and asymmetrical sensorineural hearing loss; the former would be recommended to see an audiologist or an otolaryngologist who works with an audiogist, the latter would be recommended to an ENT clinical specialist. The HHC Pro software compiles the hearing information previously available to physicians, classifies the data as normal or abnormal with respect to any measureable hearing loss, and identifies the presence of asymmetric hearing results. dB Diagnostics has chosen to validate the algorithm determining the difference between the symmetry and asymmetry of hearing loss as a greater that 20 dB difference between two ears at any test frequency (which are 250, 500, 1000, 2000, 6000, and 8000 Hz). The health care provider uses the HHC Pro to enter hand-held tympanometry results, limited clinical history, and limited physical examination findings, including a Rinne tuning fork test at 512Hz, to further identify patients who would benefit from an ENT evaluation. The HHC Pro device comprises the HHC Pro Software. With the HHC Pro Software, dB Diagnostic Systems distributes the Amplitude T4 audiometer or the OTOpod audiometer, both manufactured by Auditdata A/S and collectively known as a "digital audiometer", a 512Hz Rinne tuning fork, and the Microsoft Surface Pro tablet preloaded with the Microsoft Windows 10 operating system and the HHC Pro Software. An Amplivox Otowave 102 Tympanometer may also be distributed with the HHC Pro software.

The provided text describes a 510(k) submission for the "Hearing Healthcare Pro" software. It focuses on demonstrating substantial equivalence to a predicate device rather than presenting a detailed de novo clinical study with specific acceptance criteria and detailed performance data often found in novel device approvals.

Therefore, much of the requested information regarding acceptance criteria and a study proving the device meets them (especially in the context of typical AI/ML medical device studies with large datasets, expert ground truth, and MRMC studies) is not explicitly available or detailed in this document. The document describes software validation and verification against internal performance specifications and established standards, rather than a clinical performance study with defined acceptance criteria for AI algorithms.

However, based on the information provided, here's what can be extracted and inferred:

1. A table of acceptance criteria and the reported device performance:

The document does not provide a quantitative acceptance criteria table or specific performance metrics like sensitivity, specificity, accuracy, or AUC for an "abnormal" determination or an "asymmetry" detection. Instead, it states:

2. Sample size used for the test set and the data provenance:

• Sample Size: Not explicitly stated. The document refers to "software validation bench testing" where "HHC Pro input test results" were compared. This implies a test set, but its size (e.g., number of patient cases, number of data points) is not quantified.
• Data Provenance: Not specified regarding country of origin or whether it was retrospective/prospective clinical data or simulated/bench data. The term "input test results" suggests controlled inputs for testing the software's logic.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• Number of Experts: Not specified. It refers to "Board-Certified Otolaryngologist reviewers" (plural, so at least two).
• Qualifications: "Board-Certified Otolaryngologist". No mention of years of experience.

4. Adjudication method for the test set:

• Not explicitly stated. The document mentions "Board-Certified Otolaryngologist reviewers for equivalence in the output responses." This suggests their output was considered the "gold standard," but the method of reconciliation if there were disagreements among reviewers (e.g., 2+1, consensus, majority vote) is not detailed.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

• No, an MRMC comparative effectiveness study was not done or reported. The device is described as "compiling the hearing information previously available to physicians, classifies the data as normal or abnormal with respect to any measureable hearing loss, and identifies the presence of asymmetric hearing results." It's a screening tool that summarizes inputted data and provides a determination/recommendation, not an AI to assist human image interpretation that would typically warrant an MRMC study.

6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:

• The study described is essentially a standalone (algorithm only) validation in the sense that the software's classification outputs ("normal"/"abnormal" determination, "asymmetry" detection) were compared to the "gold standard" established by Otolaryngologist reviewers. The "inputs" to the software are measurements and observations (like audiometry results, tympanometry results, physical examination findings, patient history) provided by a human user, meaning it's a computational device processing structured input data to produce an output, rather than interpreting raw medical images or signals directly. The validation appears to focus on the correctness of this processing logic.

7. The type of ground truth used:

• Expert Consensus / Expert Review: The ground truth for the software's output (e.g., "normal" or "abnormal" determination, "asymmetry" identification) was established by "Board-Certified Otolaryngologist reviewers."

8. The sample size for the training set:

• Not applicable / Not specified. This device is presented as a software that processes user-inputted audiometric, tympanometric, and clinical data based on predefined rules (e.g., >20dB difference for asymmetry), rather than a machine learning model that undergoes a distinct "training phase" on a dataset. The document makes no mention of machine learning, neural networks, or deep learning, which would typically involve a training set. It appears to be a rule-based system or an algorithm with pre-defined thresholds.

9. How the ground truth for the training set was established:

• Not applicable. As implied above, there's no mention of a "training set" in the context of machine learning. If the "algorithm" refers to pre-defined rules, those rules would have been established by clinical knowledge and criteria, not by a data-driven training process. The validation described is focused on verifying the correct application of these rules against expert review.

Ask a specific question about this device

The Hearing Healthcare Pro software will capture audiometry test results, and will allow user input of tympanometry results, physical examination results, and patient history. The software will summarize the inputted data and provide a "normal" or "abnormal" determination. If the decision is "abnormal", the software will make a recommendation for clinical follow-up. The dB Diagnostic Systems, Inc. software is for adult use only and is intended to be used by healthcare providers not normally trained or experienced in hearing, audiology, or otology, such as primary care providers.

The Hearing Healthcare Pro software is not intended to make a clinical diagnosis.

The Hearing Healthcare (HHC) Pro is software designed to be used by healthcare providers (HCP) not normally trained or experienced in hearing, audiology, or otology, such as primary care providers. The HHC Pro software is not intended to make a clinical diagnosis. The HHC Pro software is a screening product. The software aggregates the inputted data and determines whether the person's hearing is considered normal or abnormal. If the results are abnormal, the software provides a recommendation for further clinical consultation with a licensed physician with expertise in Ear, Nose and Throat (ENT).

The HHC Pro software compiles the hearing information previously available to physicians, classifies the data as normal or abnormal with respect to any measureable hearing loss, and identifies the presence of asymmetric hearing results. The health care provider uses the HHC Pro to enter hand-held tympanometry results, limited clinical history, and limited physical examination findings, including a Rinne tuning fork test at 512Hz, to further identify patients who would benefit from an ENT evaluation.

The HHC Pro device comprises the HHC Pro Software. With the HHC Pro Software. dB Diagnostic Systems distributes the Auditdata/Otovation Amplitude T4 audiometer manufactured by Auditdata A/S, a 512Hz Rinne tuning fork, and the Microsoft Surface Pro 4 tablet preloaded with the Microsoft Windows 10 operating system and the HHC Pro Software. An Amplivox Otowave 102 Tympanometer may also be distributed with the HHC Pro software.

The provided text describes the regulatory filing for the Hearing Healthcare Pro device (K171038). While it mentions performance testing and compliance with standards, it does not provide explicit acceptance criteria or a detailed study report with quantitative performance metrics for the device's determination of "normal" or "abnormal" hearing.

Here's an breakdown based on the information provided, highlighting what is present and what is missing:

1. A table of acceptance criteria and the reported device performance

Missing. The document states that the device "has been tested and is in compliance with internal software validation bench testing to compare the HHC Pro input test results to the gold standard, Board-Certified Otolaryngologist reviewers for equivalence in the output responses." However, it does not provide a table with specific acceptance criteria (e.g., sensitivity, specificity, accuracy thresholds) or the reported numerical performance values.

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

Missing. The document refers to "internal software validation bench testing" and comparing "input test results to the gold standard," but does not specify the sample size of these test cases or the provenance of the data used (e.g., number of patient records, country, retrospective/prospective nature).

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience)

The ground truth was established by "Board-Certified Otolaryngologist reviewers." The number of these experts and their specific qualifications (e.g., years of experience) are not specified.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

Missing. The document mentions "Board-Certified Otolaryngologist reviewers" establishing the gold standard, but does not describe the adjudication method used if multiple experts were involved (e.g., consensus, majority vote, independent review).

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No. A Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not explicitly mentioned or described. The device is intended to assist primary care providers who are not normally trained in audiology, but the document does not detail a study comparing their performance with and without the device. The study described is a comparison of the device's output to expert interpretation.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes. The described "internal software validation bench testing" where the "HHC Pro input test results" were compared "to the gold standard, Board-Certified Otolaryngologist reviewers for equivalence in the output responses" suggests a standalone evaluation of the algorithm's performance in making "normal" or "abnormal" determinations.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

The ground truth used was expert consensus/interpretation by "Board-Certified Otolaryngologist reviewers."

8. The sample size for the training set

Missing. The document discusses software validation for the "HHC Pro" but does not provide information on a training set or its sample size. The description points to a validation of the software's logic against expert opinion rather than a machine learning model that would require a distinct training set.

9. How the ground truth for the training set was established

Missing. As no training set is described for a machine learning model, the method for establishing its ground truth is also not provided. The document focuses on the validation of the device's output against established expert standards.

Ask a specific question about this device

The touchTymp tympanometer is an electroacoustic test instrument that produces controlled levels of test tones and signals intended for use in conducting diagnostic hearing evaluations and assisting in the diagnosis of possible otologic disorders. It features tympanometry and acoustic reflex audiometry.

The touchTymp is an auditory impedance analyser. The device is intended to change the air pressure in the external auditory canal and measure and graph the mobility characteristics of the tympanic membrane to evaluate the functional condition of the middle ear. The device is used to determine abnormalities in the mobility of the tympanic membrane due to stiffness, flaccidity, or the presence of middle ear pathologies. The device is also used to measure the acoustic reflex threshold which occurs due to contractions of the stapedial muscle following exposure to a strong stimulus. This test allow to assess between central and peripheral pathologies and to identify where the patients uncomfortable loudness level may reside. The instrument is software controlled. The software controls the probe (tone and pressure) stimuli, measures the result and presents the result on a built in display. All functions are set and interpreted by the operator. The technological characteristics are substantially equivalent with predicate device. All technological characteristics are in compliance with the consensus standard ANSI S3.39 for auditory impedance testers.

The provided text details a 510(k) premarket notification for the "touchTymp" device, an auditory impedance tester. The FDA determined the device is substantially equivalent to legally marketed predicate devices.

Here's an analysis of the acceptance criteria and study information provided:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state specific quantitative acceptance criteria or a detailed "reported device performance" table in the way one might expect for a study comparing a new algorithm's performance against a gold standard. Instead, it relies on demonstrating compliance with existing standards and substantial equivalence to a predicate device.

The closest we can get to implied "acceptance criteria" and "performance" from this document are:

Safety and EMC Compliance: The device must comply with current standards for medical device safety and electromagnetic compatibility (EMC).

Performance Equivalence: The device must perform as safely and effectively as the predicate device for its claimed purposes (conducting diagnostic hearing evaluations and assisting in the diagnosis of possible otologic disorders, featuring tympanometry and acoustic reflex audiometry).

Software Control and Interpretation: The software must effectively control probe stimuli, measure results, and present them on a display, with all functions interpretable by the operator. | Technological Characteristics Equivalence: The document explicitly states: "The technological characteristics are substantially equivalent with predicate device. All technological characteristics are in compliance with the consensus standard ANSI S3.39 for auditory impedance testers."

Safety and EMC Compliance: "The device was found in compliance with current standards [for medical device safety and EMC]."

Performance Equivalence: "The verification and validation activities show substantial equivalence with the predicate device and that the modified touchTymp is as safe and effective as the predicate device for its claimed purpose."

Software Control and Interpretation: "The instrument is software controlled. The software controls the probe (tone and pressure) stimuli, measures the result and presents the result on a built in display. All functions are set and interpreted by the operator." |

2. Sample Size Used for the Test Set and Data Provenance

The document states "Clinical tests None applicable" (page 5), indicating that no clinical test set was used for this 510(k) submission. The determination of substantial equivalence was based on nonclinical tests verifying compliance with standards and technological similarity to the predicate device.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

Not applicable, as no clinical test set was used to establish a ground truth.

4. Adjudication Method for the Test Set

Not applicable, as no clinical test set was used.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done

No. The document explicitly states "Clinical tests None applicable." This type of study would fall under clinical testing.

6. If a Standalone (i.e. algorithm only without human-in-the loop performance) was done

Yes, in a way. The "Nonclinical tests summary" discusses "design verification and validation" performed "according to current standards for medical device safety and EMC and performance of impedance tester." This implies standalone testing of the device's functional performance against engineering specifications and industry standards, without human interpretation as part of a clinical study. However, this is not performance in the sense of an AI algorithm's diagnostic accuracy.

7. The Type of Ground Truth Used

For the nonclinical tests mentioned:

• Compliance with standards: The "ground truth" was the specifications and requirements outlined in standards like ANSI S3.39 for auditory impedance testers, and general medical device safety and EMC standards.
• Equivalence to predicate device: The "ground truth" was the established performance and characteristics of the predicate device (MI 24, K905704).

8. The Sample Size for the Training Set

Not applicable. This device is not an AI/ML algorithm that undergoes a "training set." Its functionality is based on established electroacoustic principles.

9. How the Ground Truth for the Training Set Was Established

Not applicable, as there is no training set for this type of device.

Ask a specific question about this device

The Madsen Zodiac (Type 1096) is an auditory impedance tester that is intended to change the air pressure in the external auditory canal and measure and graph the mobility characteristics of the tympanic membrane to evaluate the functional condition of the middle ear. This device is also used to measure the acoustic reflex threshold and decay testing as well as eustachian tube function testing for intact and perforated tympanic membranes. (Prescription Use).

The MADSEN Zodiac Type 1096 device is designed to measure and evaluate the acoustic impedance/admittance of the human ear by means of acoustic probe signals at different frequencies and with temporal characteristics. MADSEN Zodiac is a compact device for immittance testing. You can operate Zodiac as a stand-alone unit using the built-in keypad and display, or, if you are using the OTOsuite Immittance software module, you can operate Zodiac using the PC's keyboard and mouse with the OTOsuite Immittance module acting as the display. MADSEN Zodiac supports the following probes:

• The hand-held Quick Check probe
• The two diagnostic probe types, Classic and Comfort
• A contralateral insert phone or TDH-39.

Supported tests: Depending on the configuration, Zodiac supports the following tests and functionalities:

• Tympanometry
• Reflex Screening
• Reflex Threshold
• Reflex Decay
• ETF-I (Eustachian Tube Function - Intact)
• ETF-P (Eustachian Tube Function -Perforated)
• Admittance Recording (multiple uses, e.g. patulous Eustachian Tube evaluation, acoustic reflexes with external stimulus)
• Manual Tympanometry

This is a MODIFIED version of our previous model, the predicate device.

The document describes a 510(k) premarket notification for the "Madsen Zodiac Model 1096" auditory impedance tester. The acceptance criteria and supporting study information are derived from the comparison to the predicate device (OTOflex 100 Type 1012) and the non-clinical testing performed.

Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance:

The acceptance criteria are generally established by meeting the technical specifications of the predicate device and adhering to recognized standards. The "Reported Device Performance" column reflects the specifications of the Madsen Zodiac 1096.

Ask a specific question about this device

The Interacoustics AT235 Impedance Audiometer is an electroacoustic test instrument that produces controlled levels of test tones and signals intended for use in conducting diagnostic hearing evaluations and assisting in the diagnosis of possible otologic disorders. It features tympanometry, acoustic reflex and air conduction audiometry.

AT235 is an auditory impedance analyser. The device is intended to change the air pressure in the external auditory canal and measure and graph the mobility characteristics of the tympanic membrane to evaluate the functional condition of the middle ear. The device is used to determine abnormalities in the mobility of the tympanic membrane due to stiffness, flaccidity, or the presence of middle ear pathologies. The device is also used to measure the acoustic reflex threshold which occurs due to contractions of the stapedial muscle following exposure to a strong stimulus. This test allow to assess between central and peripheral pathologies and to identify where the patients uncomfortable loudness level may reside. The uncomfortable loudness level is useful when providing rehabilitative amplification methods and determining the correct management process for the patient. The AT235 also includes basic audiometry functions. The instrument is software controlled. The software controls the probe (tone and pressure) stimuli, measures the result and presents the result on a built in display. All functions are set and interpreted by the operator (There are no interpretations of results in the device). The technological characteristics are substantially equivalent with predicate device. All technological characteristics are in compliance with the consensus standard ANSI S3.39 for auditory impedance testers

The provided document is a 510(k) summary for the Interacoustics AT235 Impedance Audiometer. This document focuses on demonstrating substantial equivalence to a predicate device, rather than providing detailed acceptance criteria and a study proving device performance against those criteria in a typical AI/software-as-a-medical-device (SaMD) context.

Therefore, many of the requested points related to acceptance criteria, specific performance metrics, ground truth establishment, training sets, and MRMC studies are not applicable to this type of regulatory submission. The submission relies on demonstrating that the new AT235 maintains the same safety and effectiveness as its previously cleared version (K994254).

Here's a breakdown based on the available information:

1. A table of acceptance criteria and the reported device performance:

This information is not explicitly stated in the document in the format of specific acceptance criteria and performance results (e.g., sensitivity, specificity, accuracy). The document states that the device was found in compliance with current standards and demonstrated substantial equivalence with the predicate device.

The technological characteristics are stated to be in compliance with the consensus standard ANSI S3.39 for auditory impedance testers. This implicitly means the device meets the performance requirements outlined in that standard, but the specific metrics and targets are not detailed in this summary.

2. Sample sized used for the test set and the data provenance:

• Test set sample size: Not applicable. This document refers to device verification and validation activities, not a study involving a test set of patient data with a defined sample size for performance evaluation in the way an AI/SaMD would.
• Data provenance: Not applicable for performance evaluation against a specific test set. The validation activities are described as "following the design control procedure."

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

Not applicable. Ground truth establishment for a patient-level test set is not described as part of this submission. The device is an electroacoustic test instrument; its performance is validated against physical and instrument standards, not typically against expert-labeled patient data in this context.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

Not applicable. There is no mention of an adjudication process for a test set of patient data.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

Not applicable. This device is an audiometer, an electroacoustic test instrument, not an AI-powered diagnostic imaging tool that would typically involve human readers interpreting results.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

Not applicable. This device is a measurement instrument. Its "performance" is its ability to accurately measure and present electroacoustic data, not to interpret or diagnose. The document explicitly states: "All functions are set and interpreted by the operator (There are no interpretations of results in the device)."

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):

The ground truth for this type of device would primarily be instrument calibration standards, physical measurements, and compliance with recognized industry standards (e.g., ANSI S3.39 for auditory impedance testers). The document states: "The device was found in compliance with current standards and demonstrated substantial equivalence with the predicate device."

8. The sample size for the training set:

Not applicable. This device does not use machine learning or AI that requires a training set of data.

9. How the ground truth for the training set was established:

Not applicable. As there is no training set for an AI/ML algorithm, there is no ground truth establishment for such a set.

Summary of what the document does state regarding verification and validation:

The "Nonclinical tests summary" section on page 4-5 clearly states:

• "Following the design control procedure the design verification and validation were performed according to current standards for medical device safety and EMC and performance of impedance tester."
• "The device was found in compliance with current standards and demonstrated substantial equivalence with the predicate device."
• "Clinical tests None applicable"

The "Conclusion" further reiterates:

• "The AT235 as a modification to the predicate device (the previous cleared revision of AT235) uses the same or identical technology and has the same intended use as the predicate device."
• "We trust that the verification and validation activities show substantial equivalence with the predicate device and that the modified AT235 is as safe and effective as the predicate device for its claimed purpose."

In essence, the "study" proving the device meets its acceptance criteria is the comprehensive design verification and validation process against relevant medical device standards (including safety, EMC, and the ANSI S3.39 performance standard) to demonstrate substantial equivalence to its predicate. The specifics of these tests (e.g., specific measurement ranges, accuracy tolerances, etc.) are not detailed in this summary document but would be part of the full 510(k) submission.

Ask a specific question about this device

The Amplivox Otowave 202 is designed for use by trained operators (audiologists, general practitioners, hearing aid dispensers, and child health professionals) in hospitals, ENT clinics, and audiologist offices for the detection of possible otologic disorders in the general population associated with the functioning of the middle ear. The instrument performs two types of measurement: Tympanometry is used to measure the tympanic membrane and middle ear at a fixed frequency over a range of pressures. Reflex tests are used to measure stapedial reflexes. The Otowave measures ipsilateral reflexes and, when selected, reflex measurement is automatically carried out after a tympanogram is taken.

The Otowave 202 relies heavily on long-established practice in the clinical area of middle-ear measurements, and implements tympanometer functionality in a small desk-top device that may be powered from the mains (via a d.c. adapter) or by internal batteries. Stimuli are applied to the patient by means of a probe mounted on a flexible connecting lead for ease of use (especially with neonates) and additional contralateral stimuli are available if required via a separate plug-in transducer. Operation is via an intuitive menu-driven user interface, with test data displayed on the integral screen and capable of download to a PC via a USB connection and to a printer via an infrared communications (IrDA) interface. Tympanometers work by measuring the admittance of the tympanic membrane while the pressure in the ear canal is varied. The admittance is at a maximum when the air pressure across the tympanic membrane is balanced.

The provided document is a 510(k) premarket notification for the Amplivox Otowave 202 Portable Tympanometer. This type of submission focuses on demonstrating substantial equivalence to a legally marketed predicate device rather than undergoing a new clinical trial for efficacy. Therefore, much of the information typically found in an acceptance criteria study (e.g., sample sizes for test sets, expert-established ground truth, MRMC studies) is not applicable or not explicitly provided in the context of this regulatory submission.

Here's a breakdown based on the information available:

Acceptance Criteria and Reported Device Performance

The document presents a comparison of the Amplivox Otowave 202 to its predicate device, the Amplivox Otowave 102 (K081841). The "acceptance criteria" are implied by the specifications of the predicate device and the new device's ability to meet or exceed those specifications, as well as adherence to relevant industry standards.

Table of Acceptance Criteria and Reported Device Performance (as derived from the comparison table and testing summaries):

Study Details (Based on available information)

2. Sample size used for the test set and the data provenance:

• Sample Size: Not applicable. This submission relies on substantial equivalence through bench testing and adherence to standards, not human subject testing for performance.
• Data Provenance: Not applicable for performance testing in humans. Bench testing was performed on the device itself.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• Number of Experts: Not applicable. Ground truth for device performance was established against engineering specifications and international standards, not clinical expert consensus on patient data.
• Qualifications of Experts: Not applicable in the context of human data. The "experts" would be the engineers and regulatory specialists ensuring compliance with the technical standards and testing protocols.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

• Adjudication Method: Not applicable. This is not a study involving human evaluations or interpretations requiring adjudication.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

• MRMC Study: No. This device is not an AI-powered diagnostic tool requiring human reader studies. It's a medical measurement device.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

• Standalone Performance: Yes, in the sense that the device's technical specifications and performance against physical standards (e.g., probe tone accuracy, pressure range) were tested without human interpretation of clinical outcomes. The document states: "Bench performance testing: Measurements were carried out with an Otowave 202 tympanometer to confirm that it satisfies the requirements of ANSI S3.39-1987/IEC60645."

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):

• Type of Ground Truth: The ground truth for this device's performance is established by engineering/technical standards and specifications (e.g., IEC 60601-1, EN 60601-1-2, EN60645-5, ANSI S3.39-1987) for medical electrical equipment, EMC, and tympanometers, as well as the specifications of the predicate device.

8. The sample size for the training set:

• Sample Size for Training Set: Not applicable. This device does not employ machine learning or AI models that require a training set of data.

9. How the ground truth for the training set was established:

• Ground Truth for Training Set: Not applicable, as there is no training set for this type of device.

Ask a specific question about this device

Sentiero is a portable instrument to diagnose all ages or hearing loss. The instrument offers different test methods which can be configured to fit the professional's needs for screening or diagnostic purposes. It offers physiological test methods such as:

· Distortion Product Otoacoustic Emissions (DPOAE)

· Transient Evoked Otoacoustic Emissions (TEOAE)

· Auditory Brainstem Response (ABR)

· Auditory Steady State Response (ASSR)

· Auditory Impedance and acoustic reflex (TYMP)
Additionally if offers standard audiometry (psycho-acoustical). All physiological test methods are especially indicated for use in defining the type and configuration of hearing loss particularly for individuals whose behavioral audiometric results are deemed unreliable or to assist in the diagnosis of otologic disorders. Estimation of cochlear hearing thresholds (DPTHRESH) is possible at various frequencies without the need of cooperative interaction with the patient. Acoustic reflex and tympanometry (TYMP) are featured to evaluate the function of the middle and outer ear. For each method, several protocols can be configured. The results are to be used to make further recommendations regarding appropriate intervention strategies. Therefore, Sentiero is intended for use by trained personnel such as audiologists. pediatricians. ENT doctors and other health care professionals in a medical or home environment. In the United States of America. Federal law restricts this device to sale by or on the order of a licensed physician.

Available psycho-acoustical methods on Sentiero are especially indicated for use with cooperative patients starting at the age of 2 years or adequate development age, which enables them to do play/interactive audiometry. All other modules are suitable to be used for all ages elder than.infants from 34 weeks (gestational age) that are ready for discharge from the hospital.

Sentiero is designed for:

1. Diagnostics, monitoring and follow-up after newborn hearing screening
2. Pre-school, school, and adult hearing screening
3. ENT diagnostics based on measurement of
   a) Otoacoustic emissions
   b) Tympanometry and acoustic reflex
   c) Auditory Brainstem Responses
   d) Auditory Steady State Responses

Sentiero must not be used in cases of external otitis (outer ear canal infection) or in any case which yields to pain when inserting the ear probe or applying any other transducer.

Sentiero is an audiometric examination platform which consists of the Sentiero device with a touch screen display together with different accessories such as mains adapter, OAE probes, headphones, bone conductor, electrode cable, ear coupler cable, patient response switch. All connectors and transducers have a special plug in order to ensure the correct connection to the device. All plugs of the transducers have a memory chip inside which stores the information about the respective transducer (including type of connector, calibration table). As a result, the Sentiero instrument can be connected flexibly to different kind of transducers while enabling the instrument to 'know' the features of the connected transducer. This information is used within the different modules (test methods which are configured) to guide the user (feedback via display) and help to ensure correct performance of the test methods.

Sentiero is available in two different models: handheld or desktop version. The desktop version is labeled Sentiero (Type Desktop). Both versions are standalone examination platforms and can be connected to a personal computer (PC) via USB for data review and management. The handheld version is portable and is meant to be mainly used as mobile device. The desktop version is portable as well but is meant to be mainly used as stationary device. Both models base on a common hardware platform (printed circuit board, PCB) but with different configurations. Materials in contact with humans are selected to be biocompatible.

Furthermore, each model can be configured to allow different test methods and features (modules) by a license key in the device. Sentiero is based on configurable modules. Sentiero can have one single module or a combination of multiple of modules described in the following intended use.

The measurement application is controlled from a self-contained firmware (software installed on the instrument). The measurement flow is menu guided on a touch screen. Evaluation of test results is based on signal statistics (if available for the test method). Besides that wave forms and result information is displayed for the user's evaluation.

Here's a breakdown of the acceptance criteria and the study information for the Sentiero device, based on the provided 510(k) summary:

Acceptance Criteria and Device Performance for Sentiero

The provided document describes the Sentiero device and its comparison to predicate devices, primarily focusing on its substantial equivalence through adherence to standards and qualitative clinical evaluations. There are no explicit, quantifiable acceptance criteria tables with specific thresholds and reported performance metrics as one might typically find for an AI/ML device.

Instead, the acceptance criteria are implicitly defined by compliance with a comprehensive set of national and international standards related to audiological devices, electrical safety, EMC, software lifecycle, and biocompatibility, as well as qualitative clinical equivalence to predicate devices demonstrated through user feedback and expert evaluation.

1. Table of Acceptance Criteria and Reported Device Performance

Given the lack of explicit performance metrics for comparison against acceptance criteria in the provided text for the Sentiero device itself, the table below will summarize the types of criteria and the manner in which performance was assessed and deemed acceptable, primarily through equivalence to predicate devices and compliance with standards.

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Size: For the clinical evaluations, a "minimum number of participants per evaluation was 20." This applies to the external user validation studies for various modules (PTA, MAGIC, MATCH, DPTHRES, ABR, TEOAE, DPQUICK, ASSR).
  • For the ABR clinical evaluation comparing Sentiero and Chartr EP 200, the sample size is not explicitly stated but implies "the same test persons."
  • For the in-house validation using a simulator device, the "test protocol is attached in the Appendix" but not provided.
• Data Provenance: The external clinical evaluations involved various institutions, predominantly in Europe (Germany, Italy, Switzerland).
  • Center of Communication Disorders, ENT Clinic, University Medical Center of the Johannes-Guttenberg-University Mainz, Mainz, Germany
  • Department of Audiology and Speech Therapy, University of Ferrara, Ferrara, Italy
  • ENT Clinic, Klinikum rechts der Isar, Technische Universität München, Munich, Germany
  • ENT Clinic, Kantonsspital Liestal, Liestal, Switzerland
  • Department of Phoniatry and Pedaudiology, ENT Clinic, St. Elisabeth-Hospital, Bochum, Germany
• Retrospective or Prospective: The text implies prospective clinical evaluations where external users "were asked to evaluate the performance" and "Comparison measurements had to be done." The "first clinical evaluation in 2011" for ABR also suggests a prospective study explicitly comparing the two devices.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications

• Number of Experts: For the ABR clinical evaluation (Sentiero vs. Chartr EP 200), "All results have been evaluated by an expert in audiology." While only one explicit expert is mentioned for this specific evaluation, the list of external testers includes multiple individuals with titles like "Prof. Dr. med." and "Ph.D.", suggesting multiple expert evaluators across the various module assessments.
• Qualifications of Experts:
  • "expert in audiology" (for ABR evaluation)
  • Prof. Dr. med. Annerose Keilmann
  • Ph.D. Stavrous Hatzopoulos (Department of Audiology and Speech Therapy)
  • Prof. Dr. med. Hans Peter Niedermeyer (ENT Clinic)
  • Dr. Thomas Rosner (ENT Clinic)
  • PD Dr. med. Nicolas Schmuziger (ENT Clinic)
  • Prof. Dr. Katrin Neumann (Department of Phoniatry and Pedaudiology. ENT Clinic)
  • These qualifications (Prof. Dr. med., Ph.D., associated with university medical centers and ENT clinics) indicate highly qualified professionals in audiology, ENT, and related fields.

4. Adjudication Method for the Test Set

The document does not explicitly describe an adjudication method like 2+1 or 3+1 for resolving discrepancies among multiple readers or interpretations of results. The clinical evaluations suggest expert review and comparison to predicate devices, where the "results have been evaluated by an expert" or "out of these studies reports were generated or published, which indicated that Sentiero is an equivalence to the given predicate devices." This implies a form of expert consensus or direct comparison, rather than a formal adjudication process for disagreements.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No formal MRMC comparative effectiveness study, as typically understood in the context of AI performance improvement, is described. The clinical evaluations involved "multiple users" to verify usability and comparison measurements with predicate devices. However, the goal was equivalence in clinical application and usage, not quantifying an "effect size of how much human readers improve with AI vs without AI assistance" because the Sentiero device is not an AI assistant for human readers; it's a diagnostic instrument. The "AI" (advanced algorithms/signal processing) is embedded within the device's diagnostic output, rather than providing assistance to a human making a read.

6. Standalone Performance Study (Algorithm Only)

The Sentiero device is presented as a "standalone examination platform" for measurements. The document indicates that "The measurement application is controlled from a self-contained firmware (software installed on the instrument)." The clinical evaluations and in-house validations assessed the device as a whole, including its algorithms.

• "Performance tests were also conducted on a test bench. For this reason a simulator device, which simulates TEOAE, DPOAE or ABR output was used for proof of validity of the derived signal." This simulator was used to validate the performance of the test modules, which implies assessing the algorithms' ability to correctly process simulated signals and produce valid results, thus serving as a form of standalone testing for the core algorithmic functionality.
• The "signal processing information is used in Sentiero only as an indicator (traffic light or indicator for valid response)." This suggests the internal algorithms generate definitive outputs (pass/refer type indicators, waveform analysis) that are then interpreted by the user.

7. Type of Ground Truth Used

The ground truth used for the evaluations appears to be a combination:

• Expert Consensus/Clinical Judgment: The ABR evaluation states an "expert in audiology" evaluated results. The external testers, who are medical professionals and researchers, made judgments on equivalence based on clinical application/usage.
• Comparison to Predicate Devices: Clinical evaluations explicitly involved comparison measurements with "predicate devices" on "normal hearing patients with hearing loss." The predicate devices represent the established standard of care and thus serve as a form of ground truth or reference for demonstrating equivalence.
• Simulator Data: For in-house validation, a "simulator device" delivered "noise or dedicated normative answers derived from literature." Here, the "normative answers" serve as a controlled, known ground truth for validating the device's signal processing.
• Established Algorithms/Literature: "all implemented modules on Sentiero are derived from well-known algorithms, which are described in detail in published literature... The design process of Sentiero included in-house verification tests also with respect to the mentioned literature and and establishment of normative data." This means the foundational correctness of the algorithms is rooted in established audiology science.

8. Sample Size for the Training Set

The document does not mention a "training set" in the context of modern machine learning that would require a specific sample size. The Sentiero device's modules are described as being based on "well-known algorithms" and "normative data," rather than being "trained" on a dataset in the AI/ML sense. The development process involved "in-house verification tests" and "establishment of normative data," which implies calibration and validation against known physiological responses, but not a distinct training phase with a dedicated dataset as might be found in deep learning.

9. How Ground Truth for the Training Set Was Established

As noted above, there is no explicit "training set" described for the Sentiero device in the AI/ML sense. The algorithms are based on established audiological science. The "normative data" and "well-known algorithms" from literature effectively serve as the "ground truth" for the device's inherent design and calibration. This would have been established through years of audiology research and clinical practice, rather than a specific set of labeled data for this device.

Ask a specific question about this device

The Inventis Middle Ear Analyzers, Flute and Viola, are intended for use by trained operators in hospitals, nurseries, ENT clinics and audiology offices in conducting diagnostic hearing evaluations and assisting in diagnosis of possible otologic disorders. The Inventis Middle Ear Analyzers are intended for use on individuals six months of age or older. Flute is a tympanometer. Viola is a combination of audiometer and tympanometer, which reduces the amount of necessary equipment.

Flute is stand-alone tympanometry / reflex diagnostic device. Viola is a stand-alone combined unit, integrating a screening middle ear analyzer and a diagnostic audiometer capable of performing air and bone conduction threshold and speech exams. Flute and Viola are available with or without the integrated thermal printer.

Here's an analysis of the acceptance criteria and supporting study for the Inventis Middle Ear Analyzers, based on the provided 510(k) summary:

The document indicates that the Inventis Middle Ear Analyzers (Flute and Viola) underwent bench testing to demonstrate substantial equivalence to a predicate device, the Interacoustics Audiotraveller, Model AA222 (K022728). The acceptance criteria are largely implicitly defined by compliance with established electrical, mechanical, electromagnetic compatibility, and performance standards for auditory impedance testers and audiometers.

1. Table of Acceptance Criteria and Reported Device Performance:

2. Sample Size Used for the Test Set and Data Provenance:

The document describes bench testing, which typically does not involve human subjects or a "test set" in the sense of clinical data. It refers to testing on the device itself against engineering and performance standards. Therefore, specific sample sizes for a "test set" or data provenance (country of origin, retrospective/prospective) are not applicable in the context of this 510(k) submission, as it focuses on demonstrating technical equivalence rather than clinical performance from patient data.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:

Not applicable. Ground truth for bench testing is established by the specifications defined in the relevant engineering and performance standards (e.g., ANSI S3.39, IEC 60601-1). There is no "test set" requiring expert adjudication for diagnosis, as would be the case in a clinical study for an AI-powered diagnostic device.

4. Adjudication Method for the Test Set:

Not applicable, as there is no clinical "test set" in this context that would require adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done:

No. An MRMC comparative effectiveness study was not done. This 510(k) submission is for a traditional medical device (auditory impedance tester and audiometer), not an AI-assisted diagnostic tool.

6. If a Standalone (algorithm only without human-in-the-loop performance) was done:

Not applicable. The device described is a medical instrument for conducting diagnostic evaluations, not an algorithm, and its performance is inherently mechanical/electronic, not based on an independent algorithm's output.

7. The type of ground truth used:

The ground truth for the "tests" performed (bench testing) is defined by the specifications and criteria outlined in the referenced international and national standards (e.g., ANSI S3.39, IEC 60645-5, IEC 60601-1, IEC 60601-1-2, ANSI S3.6, IEC 60645-1). These standards set the benchmarks for electrical safety, EMC, and the accuracy/performance parameters of impedance and audiometry measurements.

8. The sample size for the training set:

Not applicable. This device is not an AI/machine learning device that requires a training set.

9. How the ground truth for the training set was established:

Not applicable. As no training set is used, no ground truth needed to be established for it.

Ask a specific question about this device

The Titan Impedance System is an electroacoustic test instrument that produces controlled levels of test tones and signals intended for use in conduction diagnostic hearing evaluations and assisting in the diagnosis of possible otologic disorders. It features tympanometry and acoustic reflexes.

The Titan Impedance System measures various acoustic properties of the ear, namely power reflectance, power absorption, transmittance, reflectance group delay, complex acoustic impedance and admittance, and equivalent ear canal volume. These measures allow the evaluation of the functional condition of the middle and outer ear.

The Titan Impedance System is suitable for all populations including new-born infants.

The Titan Impedance System is to be used by trained personnel only such as audiologists, ENT surgeons, doctors, hearing healthcare professionals or personnel with a similar level of education. The device should not be used without the necessary knowledge and training to understand its use and how results should be interpreted.

The instrument is audiometric equipment used for measuring aural acoustic impedance and admittance (According to ANSI S3.39).

The Titan is a platform with multi functions depending of licensing. The license function for this submission is IMP440 (IMP) for measurement of aural acoustic impedance and admittance.

The Titan IMP test is intended to change the air pressure in the external auditory canal and measure and graph the mobility characteristics of the tympanic membrane to evaluate the functional condition of the middle ear. The IMP test function is already cleared by FDA 510(k) premarket notification K083861 (04/09/2009)

The instrument is a diagnostic/clinical instrument (as defined in IEC60645-5), but provides normative data for guidance.

Titan consists of a handheld unit named Titan, Titan Cradle (cleared by FDA 510(k) premarket notification K083861) and PC software (Titan Suite/IMP440). The measurements are controlled by the handheld unit.

A connection box (shoulder box) enables different types of accessories to be connected to the platform

The Titan platform can connect to PC software via a Bluetooth connection.

The Interacoustics Titan IMP440 WBT is an electroacoustic test instrument intended for diagnostic hearing evaluations and assisting in the diagnosis of otologic disorders. It measures various acoustic properties of the ear, including power reflectance, power absorption, transmittance, reflectance group delay, complex acoustic impedance and admittance, and equivalent ear canal volume.

Here's an analysis of the acceptance criteria and the study proving the device meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The provided 510(k) summary focuses on demonstrating substantial equivalence to predicate devices rather than detailing specific numeric acceptance criteria for performance metrics. The comparison table (on pages 3-5 of the original document) primarily highlights similarities and differences in technical characteristics and functionalities between the Titan IMP440 WBT and its predicate devices (Titan with IMP440 and MIMOSA HEARID).

Implicit Acceptance Criteria (based on substantial equivalence):

2. Sample Size Used for the Test Set and Data Provenance

The document explicitly states: "No clinical tests were performed" (page 9) for the Titan IMP440 WBT device in this 510(k) submission. Therefore, there is no test set sample size mentioned, nor any data provenance (country of origin, retrospective/prospective). The assessment relies on compliance with international standards and comparison to predicate devices' performance.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

Given that no clinical tests were performed, no experts were used to establish ground truth for a test set as part of this specific submission.

4. Adjudication Method for the Test Set

Since no clinical tests and no test set were used, there was no adjudication method applied for this submission.

5. Multi-Reader, Multi-Case (MRMC) Comparative Effectiveness Study

No MRMC comparative effectiveness study was done. The submission solely focuses on demonstrating substantial equivalence through technical comparisons and compliance with standards, not on human reader performance improvement with or without AI assistance.

6. Standalone (Algorithm Only) Performance Study

No standalone (algorithm only) performance study was conducted that would typically involve a test set. Instead, the device's functional performance was verified through compliance with recognized standards (IEC 60645-5 / ANSI S3.39) and comparison against the technical specifications of predicate devices.

7. Type of Ground Truth Used

For the newly submitted device, no direct ground truth (e.g., pathology, outcomes data) was established for its own performance. The grounds for acceptance criteria are:

• Compliance with international standards: IEC 60601-1, IEC 60601-1-2, IEC 60645-5 / ANSI S3.39.
• Substantial equivalence to predicate devices: The functionality and performance characteristics of the Titan IMP440 WBT were compared to the cleared Titan IMP440 and MIMOSA HEARID. The assumption is that if it is substantially equivalent and meets standards, it is safe and effective.

The normative data included in the device (mentioned as "present" in the comparison table) are based on published research [1, 2], which would inherently have involved clinical studies to establish their "ground truth" (e.g., healthy ear function). However, this is for the normative data itself, not for a direct performance study of the Titan IMP440 WBT in this submission.

8. Sample Size for the Training Set

As the device described is a medical instrument for measurement and diagnosis (not an AI algorithm that "learns" from data in a conventional sense), the concept of a "training set" is not applicable in the way it is for machine learning algorithms. The device's functionality is engineered based on established acoustic principles and standards. The "normative data" could be considered analogous to a statistical reference for interpretation rather than a training set for an algorithm. However, the sample sizes for the studies used to generate these normative data [1, 2] are not provided in this excerpt, though they would have been part of the original research.

9. How the Ground Truth for the Training Set Was Established

Again, the concept of a "training set ground truth" is not directly applicable. If considering the normative data as an analogous reference, its "ground truth" would have been established through the research studies it was based on (e.g., "adults with normal hearing" [1], and "newborn infants" [2]). These studies would have involved clinical assessments to define "normal" (or the specific conditions being studied). However, the details of these ground truth establishments are beyond the scope of this 510(k) summary, referring instead to external academic publications.

Ask a specific question about this device

The ER35 ERO•SCAN Pro Hearing Test System is indicated for testing of cochlear and middle ear function in infants, children, and adults by measuring otoacoustic emissions (OAEs), tympanometry, and acoustic reflex.

The ER35 is a microprocessor-controlled instrument designed to screen otoacoustic emissions, tympanic membrane performance, and acoustic reflex. Test information is stored in memory, displayed on a graphic LCD, and can be printed by a printer or stored on a computer. The product is manufactured and delivered completely assembled to the retailer using materials and techniques widely used by manufacturers of hearing devices.

The provided text does not contain specific acceptance criteria or a detailed study proving the device meets said criteria in the format requested. Instead, it describes the ER35 ERO-SCAN Pro Hearing Test System, its intended use, and indicates that performance testing was conducted to support its substantial equivalence to predicate devices.

Here's an analysis of what can be extracted and what information is missing based on your request:

1. A table of acceptance criteria and the reported device performance

• Acceptance Criteria: Not explicitly stated as specific numerical targets for performance metrics (e.g., sensitivity, specificity, accuracy). The document states that "Certain performance specifications may vary compared to predicate devices, but in these cases, sufficient performance testing has been done to support substantial equivalence determination." This implies the acceptance criteria were based on demonstrating comparable performance to predicate devices, rather than predefined quantitative thresholds.
• Reported Device Performance: Not provided in the document. The text only says "Performance data collected supports a substantial equivalence determination."

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Sample Size (Test Set): Not specified. The document only mentions "Subjects tested included both male and female participants, an appropriate age range, and variable audiometric profiles (i.e., subjects with normal range hearing and impaired hearing function)."
• Data Provenance: Not specified. It's unclear if the study was prospective or retrospective, or the country of origin of the data.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

• Not specified. The document does not describe how ground truth was established, nor the involvement or qualifications of experts for this purpose.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

• Not specified.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

• MRMC Study: Not mentioned or applicable. This device is an audiometer, not an AI diagnostic tool primarily requiring human interpretation of images or complex data where AI assistance would be a factor. The "study" referenced is performance testing of the device itself.
• Effect size of human readers with/without AI assistance: Not applicable to this device.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

• Yes, a standalone performance assessment was done. The "performance testing" referenced pertains to the ER35 device itself, which performs measurements of otoacoustic emissions, tympanometry, and acoustic reflex. This is inherently a standalone assessment of the device's measurement capabilities. The text states: "Non-clinical and clinical validation was performed in accordance with recognized consensus standards for this type of device. The ER35 has been demonstrated to be in compliance with the following performance and safety standards..."

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

• Not explicitly stated. Given the nature of the device (measuring cochlear and middle ear function), the "ground truth" would likely be established through recognized clinical diagnostic methods or reference standards for otoacoustic emissions, tympanometry, and acoustic reflex, possibly involving expert clinical assessment or other validated diagnostic equipment. The document mentions "variable audiometric profiles (i.e., subjects with normal range hearing and impaired hearing function)," implying clinical classifications were used.

8. The sample size for the training set

• Not applicable as this is not an AI/machine learning device that requires a training set. The device is a physical instrument for measurement.

9. How the ground truth for the training set was established

• Not applicable for the same reason as point 8.

Summary of what is available and what is missing:

The document describes the submission as a 510(k) for an updated medical device (an audiometer) demonstrating substantial equivalence to predicate devices. The "study" mentioned is "performance testing" conducted to show compliance with medical device standards (ANSI S3.39-1987 (R2007), IEC 60601-1:2000, ANSI S3.6-2004, IEC 60645-5:2004) and to support substantial equivalence.

It does not provide the granular detail requested common for AI/ML device evaluations, such as specific performance metrics (sensitivity, specificity, AUROC), sample sizes for test sets, details on ground truth establishment by experts, or MRMC studies. The key statement regarding performance is: "Certain performance specifications may vary compared to predicate devices, but in these cases, sufficient performance testing has been done to support substantial equivalence determination. Non-clinical and clinical validation was performed in accordance with recognized consensus standards for this type of device."

Ask a specific question about this device