Search Results

The EPAD 2 system is intended for use in monitoring neurological status by recording somatosensory evoked potentials (SSEP), electromyography (EMG), or assessing the neuromuscular junction (NMJ). Neuromonitoring procedures include intracranial, extracranial, intratemporal, extratemporal, neck dissections, upper and lower extremities, spinal degenerative treatments, pedicle screw fixation, intervertebral fusion cages, rhizotomy, orthopedic surgery, open or percutaneous, lumbar, thoracic, and cervical surgical procedures.

The EPAD 2 System consists of the EPAD 2 Headbox and power supply, a tablet computer, electrodes, stimulating probe or clip and cables. The Headbox is responsible for the stimulation and acquisition of signals for the intraoperative neurophysiologic monitoring (IONM) functions. The Headbox contains onboard firmware and is mounted on the operating room (OR) bed by clipping it to the bed side bar. The EPAD application provides the primary graphical user interface and controls for the EPAD II System. The application runs on a touchscreen tablet mobile device which connects wirelessly or via a wired USB cable to the Headbox. The application serves as the interface to the EPAD Headbox, enabling both user input (e.g., patient and procedure information, adjustment of stimulus and acquisition parameters) and display of output (e.g., display of acquired waveforms, data, messages and alerts to the clinician).

The provided text primarily focuses on the 510(k) clearance of the EPAD 2 System, demonstrating its substantial equivalence to predicate devices rather than presenting data from a specific study designed to meet acceptance criteria for novel performance claims. Therefore, much of the requested information regarding acceptance criteria, reported performance, and study details is not explicitly available in the provided document.

However, I can extract information related to the device's characteristics and the types of testing performed to support its substantial equivalence.

Here's an analysis based on the given document:

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

The document does not explicitly present a table of acceptance criteria with corresponding reported device performance metrics in the way one would expect for a study proving new performance claims. Instead, it relies on demonstrating substantial equivalence to predicate devices. The performance data mentioned are related to meeting functional, system, and software requirements, and compliance with safety standards.

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

The document states: "Not applicable; determination of substantial equivalence is not based on an assessment of clinical performance data."
This indicates that:

• There was no clinical test set in the traditional sense for evaluating novel performance claims against acceptance criteria.
• The data provenance, sample size, or retrospective/prospective nature of a clinical study are not specified because such a study was not the basis for clearance.
• The performance data cited are from nonclinical performance testing, EMC, electrical safety, and usability testing, which typically involve device-level assessments rather than patient-based test sets.

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)

Since there was no clinical test set for evaluating novel performance claims, there is no information provided regarding experts establishing ground truth for such a test set. The substantial equivalence determination relies on comparisons to already cleared predicate devices.

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

Not applicable, as no clinical test set requiring ground truth adjudication for novel performance claims was used as the basis for this 510(k) clearance.

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. The EPAD 2 System is an Evoked Response Electrical Stimulator used for neuromonitoring, not an AI-assisted diagnostic imaging device that would typically involve a multi-reader multi-case study or "human readers improve with AI" metrics.

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

This concept is not directly applicable to the EPAD 2 System. The device is an instrument used by qualified medical personnel for neuromonitoring and does not operate as a standalone "algorithm only" device for making diagnoses or performing interventions without human-in-the-loop interaction. Its function is to acquire and display physiological signals for interpretation by a clinician.

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

For the nonclinical performance, EMC, electrical safety, and usability testing, the "ground truth" would be established by:

• Engineering specifications and regulatory standards: For functional, system, and software requirements, and for EMC/electrical safety, the device's performance is compared against pre-defined engineering specifications and national/international standards (e.g., IEC 60601-1).
• User feedback/observation: For usability testing, the "ground truth" is typically the observation of user interaction and compliance with usability engineering principles, identifying potential use errors or safety concerns.

There is no mention of pathology or outcomes data as ground truth for this 510(k) submission, as it explicitly states that clearance is "not based on an assessment of clinical performance data."

8. The sample size for the training set

Not applicable. This device is cleared through the 510(k) pathway which demonstrates substantial equivalence to predicate devices. It is not an AI/ML device that typically requires a large "training set" in the machine learning sense for its core functionality. While there might be firmware/software development which internally used data, the submission does not detail such a training set for regulatory review.

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

Not applicable, as no training set in the AI/ML context is described or relevant for this 510(k) cleared device.

Ask a specific question about this device

The EPAD is intended for use in monitoring neurological status by recording somatosensory evoked potentials (SSEP) or assessing the neuromuscular junction (NMJ).

The EPAD system consists of the following components/accessories:

EPAD Headbox EPAD Headbox Power Supply EPAD Tablet Computer (includes power supply and USB cable) Stimulator Left Blue Cable Assembly (110 inches) Stimulator Right Yellow Cable Assembly (110 inches) Acquisition Left Red Cable Assembly (91 inches) Acquisition Right White Cable Assembly (92 inches) Stimulator Left Blue Short Cable Assembly (67 inches) Stimulator Right Yellow Short Cable. Assembly (67 inches) Acquisition Left Red Short Cable Assembly (44 inches) Acquisition Right White Short Cable Assembly (72 inches) Adapter Cable for EPAD Headbox (for leakage current testing)

Upper Limb Electrodes package

Lower Limb Electrodes package

The EPAD Headbox contains a complete data acquisition system that has built-in amplifiers, analog to digital converters, and digital signal processors. User interface is via tablet touchscreen computer provided with the EPAD System and running the Android operating system. The EPAD software application is preloaded onto the tablet. Data can be transferred to an external computer for archiving purposes. Communication between the EPAD Headbox and tablet is via Bluetooth wireless or USB connection.

Electrode cables are provided for left and right stimulation and left and right acquisition, color coded for correct connection to the EPAD Headbox: Custom cutaneous electrodes for stimulus and acquisition are provided by SafeOp for use with the EPAD. A total of 11 electrodes are applied for full patient monitoring (upper and lower limbs). The SafeOp electrodes are wet qel, single patient use, disposable, and biocompatible for short term (

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

1. Table of Acceptance Criteria and Reported Device Performance

The 510(k) summary for the EPAD™ primarily establishes substantial equivalence through a comparison with predicate devices and nonclinical testing that confirms safety and performance against established standards. Direct "acceptance criteria" for diagnostic performance (like sensitivity/specificity) are not explicitly stated for a clinical study comparing the AI's diagnostic ability to a gold standard. Instead, the acceptance is based on meeting technical specifications and demonstrating safety and effectiveness.

However, we can infer performance criteria from the non-clinical testing performed and the comparison to predicate devices. The "reported device performance" refers to the results of these tests.

7. The Type of Ground Truth Used (expert consensus, pathology, outcomes data, etc.)

• For the Alert Algorithm: The "ground truth" was defined by predefined quantitative thresholds for waveform changes: >50% amplitude reduction and >10% latency increase. This is a technical, rather than a clinical, ground truth.
• For other non-clinical tests (e.g., electrical safety, biocompatibility, signal quality), the ground truth is compliance with established engineering standards and specifications.

8. The Sample Size for the Training Set

• The summary does not provide information on a training set sample size. As a 510(k) submission for a device primarily focused on evoked potential recording and basic alert logic, it doesn't describe the development of a complex machine learning model requiring a distinct training set. If the alert algorithm involved any learned parameters, this information is not detailed.

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

• Not detailed / Not applicable as a distinct training set and its ground truth establishment are not discussed in this 510(k) summary. If the alert algorithm's thresholds were derived from clinical data, the method for establishing that ground truth is not provided.

Ask a specific question about this device