Search Results

The Integrity with VEMP provides testing that is intended to assist in the assessment of vestibular function. The target patient population ranges from school age children to geriatric adults who can complete the testing tasks. The Integrity with VEMP is intended to be used by a variety of professionals, such as clinical practitioners specialized in balance disorders, Audiologists, and Otolaryngologists (ENT doctors) with prior knowledge of the medical and scientific knowledge about the VEMP procedure.

VEMP (Vestibular Evoked Myogenic Potential) is a short latency response generated either from sternocleidomastoid muscle (cVEMP) or oblique muscle (oVEMP) which is typically evoked by high level acoustic or vibratory stimulation. VEMP is a non-invasive test that is used for diagnosis of vestibular disorders such as superior canal dehiscence, Menier's disease, vestibular neuritis, and, among others. VEMP test can be done in clinics (ENT/audiology) and hospitals provided that the users have adequate knowledge and background about underlying process of VEMP and recording auditory evoked responses.

The Integrity V500 is indicated for auditory evoked potential testing. "Integrity with VEMP" (this submission) is the addition of the VEMP test modality to the Integrity V500. The Integrity V500 is an auditory evoked auditory system response and Integrity with VEMP is an auditory evoked vestibular system response. Since the vestibular system is connected to the auditory system, a loud stimulus to the auditory system also simultaneously stimulates the vestibular system. However, the evoked signals are measured at different electrode sites. It is important to note that the methods of stimulation and data acquisition are the same for the VEMP modality as for the ABR modality (one of the Integrity V500's auditory evoked testing modalities), with differences being in electrode montage and patient's physical state. As such, it is possible to also get a VEMP response while testing in the Integrity V500's ABR modality.

Integrity with VEMP is a PC based device which uses the same hardware and similar software for evoking the stimulus, collecting the response, processing the data, and displaying the outcome on the screen as does the Integrity V500. The hardware used are: a VivoLink (patient interface device), air and bone conduction transducers (to stimulate the vestibular system), CV-Amp (bio-amplifier), and a computer. The response from the muscle is picked up by a bio-amplifier attached to the neck (for cVEMP) or under eyes (for oVEMP) and forehead with electrodes. The data is the VivoLink for pre-processing and then transferred to the PC via Bluetooth connection for full processing using the algorithm designed for VEMP processing. Through the test, the processed sweeps are filtered using the filter setting defined by the user and the averaged response is shown on the screen.

The primary diagnostic component of a VEMP measurement is the comparison of the amplitude of the primary peak of the VEMP response between right and left sides, defined as an asymmetry ratio, where a significant amount of asymmetry is indicative of a vestibular disorder. The amplitude of the vestibular response peaks is also dependent on the contraction level of the muscles involved in the recording. To minimize the muscular response biasing the result, it is important to have equal contraction levels for both sides; this is especially physically challenging to generate equivalent neck contractions needed for cVEMP. To overcome this issue, two key features were added to the Integrity with VEMP compared to Integrity V500: a biofeedback EMG monitor (which displays real-time muscular activity) and VEMP response normalization (scaling based on muscular contraction levels). The EMG monitors can be used as a guidance to the clinicians and patients for the level of muscle contraction. Normalization automatically scales the recorded sweeps based on the energy of the corresponding EMG which helps to compensate for imbalance contraction from the two sides.

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

Acceptance Criteria and Reported Device Performance

The acceptance criteria are not explicitly listed as a table of thresholds to be met, but rather are inferred from the clinical studies' conclusions regarding substantial equivalence and test reliability. The reported device performance is presented in tables comparing the device's measurements to a predicate device and to literature norms, as well as its reliability metrics against industry standards.

• P1 [ms]: Integrity VEMP (15.6 ± 1.3, 15.8 ± 1.4) vs. Eclipse (14.7 ± 1.2, 15.1 ± 1.5). Both within literature norms (11-17 ms). ICC: 0.72 (Right), 0.91 (Left).
• N1 [ms]: Integrity VEMP (23.5 ± 1.6, 23.9 ± 1.5) vs. Eclipse (23.6 ± 1.4, 23.7 ± 1.8). Both within literature norms (18-27 ms). ICC: 0.95 (Right), 0.80 (Left).
• P1-N1 [µV]: Integrity VEMP (209.6 ± 105.3, 198.7 ± 106) vs. Eclipse (195.5 ± 54.4, 168.8 ± 48.8). Both within literature norms (28-300 µV). ICC: 0.67 (Right), 0.71 (Left).

oVEMP vs. Eclipse (K162037):

• N1 [ms]: Integrity VEMP (11.2 ± 0.8, 11.4 ± 0.9) vs. Eclipse (10.2 ± 1.1, 10.4 ± 1.1). Both within literature norms (9-19 ms). ICC: 0.63 (Right), 0.78 (Left).
• P1 [ms]: Integrity VEMP (15.9 ± 1.7, 16.3 ± 1.6) vs. Eclipse (15.2 ± 1.7, 15.2 ± 1.6). Both within literature norms (12-22 ms). ICC: 0.89 (Right), 0.76 (Left).
• N1-P1 [µV]: Integrity VEMP (17.4 ± 8.5, 17.3 ± 8.3) vs. Eclipse (16.7 ± 10.6, 14.1 ± 6.6). Both within literature norms (2-45 µV). ICC: 0.90 (Right), 0.91 (Left).

Conclusion: Both systems' results consistent with literature; statistical analysis shows good to excellent reliability. Substantial equivalency criteria met. | Test-Retest Reliability (One Session):

• cVEMP (Air Conduction): Mean (μ) = 0.92, Median = 0.94, Std Dev = 0.05, IQR = 0.07. (Meets industry std: Mean ≥ 0.879, IQR ≤ 0.111)
• oVEMP (Air Conduction): Mean (μ) = 0.89, Median = 0.89, Std Dev = 0.05, IQR = 0.08. (Meets industry std)
• cVEMP (Bone Conduction): Mean (μ) = 0.92, Median = 0.92, Std Dev = 0.05, IQR = 0.12. (Mean meets, IQR slightly high but deemed acceptable due to bone conduction variability)

Test-Retest Reproducibility (Two Sessions 24-72 hrs apart):

• cVEMP (Air Conduction): Mean (μ) = 0.92, Median = 0.93, Std Dev = 0.05, IQR = 0.08. (Meets industry std: Mean ≥ 0.879, IQR ≤ 0.111)
• oVEMP (Air Conduction): Mean (μ) = 0.89, Median = 0.87, Std Dev = 0.06, IQR = 0.11. (Meets industry std)

Conclusion: Reliability and reproducibility values are similar to those from other industry products (Eclipse K162037 document). Good test-retest reliability and reproducibility are concluded. |
| 2. Test-Reliability and Reproducibility comparable to industry standards | (See above) | (See above) |

1. Sample sizes used for the test set and the data provenance:

• Study 1 (Substantial Equivalence):
  • cVEMP: 13 normal adult participants.
  • oVEMP: 9 adult participants.
  • Data Provenance: Not explicitly stated (e.g., country of origin, prospective/retrospective). Implied to be prospective for the purpose of this comparative study.
• Study 2 (Reliability & Reproducibility):
  • US site: 33 normal hearing adults for cVEMP (AC); 17 normal hearing adults for oVEMP (AC).
  • Canada Site: 9 school-age normal hearing children for cVEMP (BC); 8 pre-school age normal hearing children for cVEMP (AC).
  • Data Provenance: Prospective, collected from two sites (US and Canada).

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

• The ground truth in this context is established by the physiological measurements themselves, not by expert interpretation of images or other subjective data. The "normal adult/hearing" participant selection implies a pre-qualification based on typical physiological parameters, but no specific "expert" panel is described as establishing or adjudicating individual VEMP responses for ground truth. The comparison is against established literature norms and a predicate device's measured performance.

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

• Not applicable as this is a device performance study based on objective physiological measurements (latencies, amplitudes) rather than subjective assessments requiring expert consensus or adjudication. The data analysis involved statistical comparisons (mean, standard deviation, ICC, IQR).

4. 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, this type of study was not performed. The device is not an AI-powered diagnostic tool that assists human readers in interpreting complex data like medical images. It's a medical device that measures physiological responses (VEMP). The study focuses on the device's ability to accurately capture and report these responses, which are then interpreted by clinicians.

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

• The device inherently involves a human-in-the-loop for proper patient setup, transducer placement, monitoring for muscle contraction via biofeedback, and ultimately interpreting the results. The "performance" being evaluated is the system's ability to consistently provide objective VEMP measurements. There isn't an "algorithm only" mode separate from its intended use with a human operator. The software's processing of data from the hardware is integral to its function.

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

• The ground truth is based on physiological measurement consistency and comparison to established literature norms and a legally marketed predicate device.
  • For Study 1, the ground truth for "normal" VEMP responses is referenced against "Norms from Literature" for P1/N1 latencies and P1-N1 amplitudes. The predicate device's performance also serves as a comparative "ground truth" for substantial equivalence.
  • For Study 2, "industry-accepted standards" for test-retest repeatability and reproducibility (derived from the predicate device's 510(k) document) served as the benchmark for reliability.

7. The sample size for the training set:

• This information is not provided. The document describes clinical studies that served as part of the regulatory submission (verification and validation), not studies for training a machine learning model.

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

• This information is not provided, as the studies described are related to clinical validation for regulatory clearance, not the development or training of a machine learning model. If any internal models (e.g. for artifact rejection or signal processing) were trained, their training data and ground truth establishment are not detailed in this document.

Ask a specific question about this device