Search Results

The REMI-AI Rapid Detection Module (REMI-AI RDM) is a seizure detection module which is integrated into the REMI Remote EEG Monitoring System and is only indicated for use within non-ICU (Intensive Care Unit) healthcare settings. REMI-AI RDM has not been validated for and is not indicated for detection of electrographic status epilepticus.

REMI-AI RDM conducts automated analysis of REMI EEG data in near real-time and provides notifications of potential electrographic seizures (events) through the REMI System when seizure prevalence of 10% or greater (indicating seizure activity of at least 30 seconds within a 5-minute rolling window) is detected. When seizure prevalence is displayed, the notification also displays the corresponding event detection confidence. Notifications are intended to be used by qualified clinicians who will exercise professional judgment in their application. Detected events are also annotated in the associated REMI EEG record as an aide to the qualified physician's REMI EEG review.

Delays of up to several minutes may occur between the detection of an event and the generation of an event notification, and are thus not a substitute for real-time monitoring. REMI-AI RDM does not make any diagnostic conclusion about the subject's condition and is intended as a physiological signal monitor. REMI-AI RDM is indicated for use with adult and pediatric patients (6+ years).

REMI-AI RDM conducts automated analysis of EEG data collected by the REMI System in near real-time. REMI-AI RDM provides notifications of the prevalence and confidence of potential electrographic seizures, having a combined prevalence of 10% or greater, which correlates with a duration of at least 30 seconds of activity within a rolling 5 minute window of EEG.

REMI-AI RDM notifications are presented through the REMI Mobile software application, and are intended to be used by qualified clinicians who will exercise professional judgment in their interpretation. Notifications include the prevalence and confidence value for the event and are marked in the associated EEG record in order to assist qualified clinicians in their assessment.

REMI-AI RDM notifications identify when a section of EEG is consistent with seizure characteristics it has been trained to recognize. When a notification is presented, clinical context and facility procedures should inform next steps in patient evaluation and management. REMI-AI RDM does not make any treatment or management recommendations.

Here's a summary of the acceptance criteria and study details for the REMI-AI Rapid Detection Module (REMI-AI RDM), based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Test Set Sample Size:
  • 22 patient records with 54 consensus-determined electrographic seizures (lasting at least 30 seconds).
  • 22 patient records with no consensus-determined electrographic seizures.
  • Total Validation Sample Size: 44 patient records.
• Data Provenance: The text does not explicitly state the country of origin. It indicates that the data was collected concurrently with standard-of-care 19-channel, full-montage, video-EEG in "Epilepsy Monitoring Units (EMUs) or for up to 3 continuous days during at-home ambulatory EEG monitoring." This suggests the data is retrospective as it was "previously acquired" for validation, but the initial data collection method (prospective/retrospective) for the source may vary.

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

• Number of Experts: 3 independent expert epileptologists for panel review, selected from a panel of 6.
• Qualifications of Experts: Certified by the American Board of Psychiatry and Neurology or certified by the American Board of Clinical Neurophysiology with Special Competency in Epilepsy Monitoring.

4. Adjudication Method for the Test Set

• Adjudication Method: Consensus ground truth was established when at least 2 of the 3 expert epileptologists agreed on the presence or absence of an electrographic seizure event (2+1).

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of Human Reader Improvement

• No, an MRMC comparative effectiveness study involving human readers with and without AI assistance was not reported in this document. The study focuses on the standalone performance of the AI algorithm.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

• Yes, a standalone performance study was done. The "Clinical Validation" section explicitly details the algorithm's performance (Sensitivity and False Alarm Rate) against a clinical reference standard, with no mention of human interaction with the AI output during this performance assessment. Notifications are "intended to be used by qualified clinicians," but the validation itself is on the algorithm's detection capabilities.

7. The Type of Ground Truth Used

• Type of Ground Truth: Expert consensus. The ground truth was established by a panel of 3 independent expert epileptologists reviewing "standard 19+channel wired 10-20 montage EEG records."

8. The Sample Size for the Training Set

• Training Set Sample Size: 117 patient records.
  • 82 patient records with seizures ("Train Sz").
  • 35 patient records without seizures ("Train No-Sz").

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

• The document states that EEG data was used to "train the REMI-AI RDM algorithm to identify potential electrographic seizure events." While it explicitly describes how the ground truth for the validation data set was established (panel review by experts), it does not explicitly detail the exact method for establishing ground truth for the entire training set. However, given the context of the validation process, it is highly probable that a similar expert review and consensus process would have been used to annotate the training data as well.

Ask a specific question about this device