(305 days)
SmartEP is an evoked response testing and diagnostic device, that is capable of eliciting, acquiring, and measuring auditory, somatosensory, visual, and vestibular evoked myogenic potential data, as well as providing nerve stimulation and monitoring.
The intended use of the SmartEP device is to objectively record evoked responses from patients of all ages upon the presentation of sensory stimuli. The product is indicated for use as a diagnostic aid and adjunctive tool in sensory related disorders (i.e., auditory, somatosensory, visual, and vestibular) and in surgical procedures for inter-operative nerve monitoring.
The SmartEP system is intended to be used by trained personnel in a hospital, nursery, clinic, audiologist's, EP technologist's, surgeon's, or physician's office, operating room, or other appropriate setting.
The SmartEP device records evoked potentials by using delivery of auditory, somatosensory, visual, or nerve sensory stimuli and using signal averaging techniques to extract the evoked potential from the uncorrelated electrical activity of the brain (electroencephalography or EEG) and muscles (electromyography or EMG). The device has options for Auditory Evoked Potentials (AEPs), Somatosensory Evoked Potentials (SEPs), Visual Evoked Potentials (VEPs), Vestibular Evoked Myogenic Potentials (VEMPs), and nerve stimulation and monitoring. The SEP, VEP, and nerve stimulation and monitoring functionality. operating principles, and intended uses are the same as on the predicate SmartEP device. On the SmartEP device with VEMP modality, the AEP modality has been modified to facilitate VEMP recording and analysis with optional biofeedback. The VEMP features added are comparable to those found in the ICS Chartr 200 predicate device. The VEMP modality does not provide a diagnosis. Diagnosis is made by a medical professional.
The SmartEP device is a Windows OS personal computer (PC) based system composed of software modules, an external main hardware unit, an optional biofeedback box, and peripheral stimulus delivery and recording components and accessories. The biofeedback box, stimulation, and recording devices are connected to the main hardware unit which is connected to the PC via a Universal Serial Bus cable. Software on the computer is used for the user interface to facilitate test parameter specification and for data display and analysis purposes.
The SmartEP with VEMP device has an optional biofeedback hardware accessory (VEMP feedback box) or uses a computer monitor for indicating EMG levels during VEMP testing. The VEMP feedback box has LEDs that indicate that the measured EMG level is either below the minimum value set by the user (Low - orange LED), or is between the minimum and maximum values set by the user (Satisfactory green LED), or is above the maximum value as set by the user (High – orange LED). The computer monitor displays a bar graph and pictorial face that indicates that the measured EMG level is either below the minimum value set by the user (Low - small pink bar and sad face), or is between the minimum and maximum values set by the user (Satisfactory - medium green bar and happy face), or is above the maximum value as set by the user (High - large pink bar and sad face). Recording of VEMPs can be set to occur when the EMG level is within the user programmed satisfactory range.
The provided document describes the substantial equivalence determination for the SmartEP device, focusing on its Vestibular Evoked Myogenic Potential (VEMP) testing modality compared to a predicate device. The information primarily addresses performance testing related to VEMP repeatability and reproducibility, rather than a typical AI model's acceptance criteria based on sensitivity/specificity/accuracy or an MRMC study.
Therefore, many of the requested fields cannot be directly answered as they pertain to AI/machine learning evaluation paradigms which are not the subject of this 510(k) summary. However, I will answer all applicable points based on the provided text.
Acceptance Criteria and Device Performance (VEMP Modality Only)
The document does not explicitly state "acceptance criteria" in terms of pre-defined numerical thresholds for performance metrics like sensitivity, specificity, or accuracy, as this is not an AI/diagnostic algorithm in the conventional sense. Instead, the performance is demonstrated through comparisons of repeatability and reproducibility metrics (specifically, correlation values of VEMP waveforms) to a legally marketed predicate device. The general acceptance criterion is demonstrating that the SmartEP device's VEMP performance is "similar or higher" than the predicate device.
Table of Performance for VEMP Waveform Correlation (SmartEP vs. Predicate)
| Performance Metric | SmartEP Device (Range of Mean Correlation) | Predicate Device (Mean Correlation) | Acceptance Criterion (Implicit) |
|---|---|---|---|
| cVEMP Waveform Test-Retest Repeatability (Session 1 Mean Correlation) | 0.911 - 0.915 | Not explicitly stated for this metric in predicate data. | Similar or higher than predicate (if available for comparison) and demonstrating high repeatability. |
| cVEMP Waveform Test-Retest Repeatability (Session 2 Mean Correlation) | 0.921 - 0.941 | Not explicitly stated for this metric in predicate data. | Similar or higher than predicate (if available for comparison) and demonstrating high repeatability. |
| oVEMP Waveform Test-Retest Repeatability (Session 1 Mean Correlation) | 0.944 - 0.956 | Not explicitly stated for this metric in predicate data. | Similar or higher than predicate (if available for comparison) and demonstrating high repeatability. |
| oVEMP Waveform Test-Retest Repeatability (Session 2 Mean Correlation) | 0.951 - 0.955 | Not explicitly stated for this metric in predicate data. | Similar or higher than predicate (if available for comparison) and demonstrating high repeatability. |
| cVEMP Waveform Reproducibility (Session 1 vs. Session 2 Mean Correlation) | 0.847 - 0.868 | 0.915 (Right), 0.916 (Left) | Similar or higher than predicate. |
| oVEMP Waveform Reproducibility (Session 1 vs. Session 2 Mean Correlation) | 0.902 - 0.940 | 0.926 (Right), 0.93 (Left) | Similar or higher than predicate. |
Reported Device Performance:
The document states: "The mean correlation values for both cVEMP and oVEMP recordings obtained the SmartEP with VEMP device for both test-retest repeatability and reproducibility are similar or higher than those reported for the ICS Chartr EP 200 with VEMP device." This generally indicates that the SmartEP device met the implicit acceptance criterion of performing comparably or better than the predicate device for VEMP testing.
Study Details
-
Sample sizes used for the test set and the data provenance:
- Study 1 (P1 latency from cVEMP): 215 normal subjects.
- Study 2 (cVEMP and oVEMP waveforms): 10 adult normal subjects.
- Data Provenance: The document does not explicitly state the country of origin. It describes the studies as "performed with using the SmartEP with VEMP device," suggesting they were conducted by or for the manufacturer. The studies appear to be prospective to evaluate the device's performance.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
This device is an evoked potential system used to record physiological responses, not an AI diagnostic algorithm that produces an interpretation requiring expert ground truth for classification/detection tasks. The "ground truth" here is the physiological response itself, as measured by the device. Therefore, this question is not directly applicable to the type of device and study described. -
Adjudication method (e.g., 2+1, 3+1, none) for the test set:
Not applicable, as the "ground truth" is the recorded physiological signal, not an interpretation requiring human adjudication. -
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. This is a medical device for measuring physiological responses, not an AI-assisted diagnostic tool that humans would use to interpret images or signals. The study focuses on the device's measurement consistency. -
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
This device is a standalone measurement system. The "performance testing data" section describes the device's inherent ability to record VEMP waveforms repeatably and reproducibly, independent of a specific human interpretation loop for a diagnostic task. -
The type of ground truth used (expert consensus, pathology, outcomes data, etc):
The "ground truth" is the physiological VEMP response itself, measured by the device. The studies characterize the device's ability to consistently and accurately capture these responses (repeatability and reproducibility). There isn't a "diagnostic ground truth" in the sense of a disease state confirmed by a gold standard like pathology, as the device is for recording diagnostic data, not making the diagnosis itself. It is stated that "The VEMP modality does not provide a diagnosis. Diagnosis is made by a medical professional." -
The sample size for the training set:
Not applicable. This is not an AI/machine learning device that requires a "training set" to learn from data. It's a measurement instrument. -
How the ground truth for the training set was established:
Not applicable, as there is no "training set."
§ 882.1900 Evoked response auditory stimulator.
(a)
Identification. An evoked response auditory stimulator is a device that produces a sound stimulus for use in evoked response measurements or electroencephalogram activation.(b)
Classification. Class II (performance standards).