Search Results

The REMI Remote EEG Monitoring System is indicated for use in healthcare settings where near real-time and/or remote EEG is warranted and in ambulatory settings where remote EEG is warranted. REMI System uses single patient, disposable, wearable sensors intended to amplify, capture, and wirelessly transmit a single channel of electrical activity of the brain for a duration up to 30 days.

REMI System uses the REMI Mobile software application that runs on qualified commercial off-the-shelf mobile computing platforms. REMI Mobile displays user setup information to trained medical professionals and provides notifications to medical professionals and ambulatory users. REMI Mobile receives and transmits data from connected REMI Sensors to the secure REMI Cloud where it is stored and prepared for review on qualified EEG viewing software.

REMI System does not make any diagnostic conclusion about the subject's condition and is intended as a physiological signal monitor. REMI System is indicated for use with adult and pediatric patients (1+ years).

The REMI System has three major components:

• 1. REMI Sensor A disposable EEG sensor which is placed on the patient's scalp using a conductive REMI Sticker
• 2. REMI Mobile A mobile medical application that is designed to run on a qualified commercial-off-the-shelf mobile computing platform (an Android tablet for use in healthcare settings, and a portable/wearable Android device (phone or smartwatch) for use in ambulatory settings), acquire EEG data transmitted from REMI Sensors and then transmit the EEG data and associated patient information via wireless encrypted transmission to.
• 3. REMI Cloud A HIPAA-compliant secure cloud storage and data processing platform where data is processed into a qualified EEG reviewing software format for neurological review.

The provided document is a 510(k) Pre-market Notification Summary for the REMI Remote EEG Monitoring System (K243185). This document details the device's characteristics, indications for use, and the studies conducted to demonstrate its substantial equivalence to a predicate device (REMI Remote EEG Monitoring System, K230933).

Based on the provided information, here's a description of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance

The document primarily relies on comparisons to its own predicate device (K230933) and general performance testing against recognized standards. Specific quantitative acceptance criteria are not explicitly detailed in a table format within this summary, but the general assertion is that the device met all predetermined acceptance criteria derived from the listed tests.

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

• Clinical Performance Test Set (Pediatric Extension):
  • Sample Size: 13 younger pediatric patients (ages 1 to <6 years)
  • Data Provenance: Retrospective review of existing REMI EEG records gathered from NIH-funded studies conducted by Epitel in support of REMI development efforts. Data collected under IRB oversight and registered under NCT03583957. The document states a single pediatric-focused site was involved.
  • Geographic Origin: Not explicitly stated, but NIH funding implies a U.S. origin.

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

• Number of Experts: One
• Qualifications: An "experienced pediatric epileptologist independent of wired EEG and generally separated in time from the actual time of collection by over a year - to ensure appropriate experience, consistency, and minimization of bias."

4. Adjudication Method for the Test Set

• Adjudication Method: None explicitly described beyond a single expert's retrospective review. The expert's review was the ground truth.

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

• MRMC Study Done? No. The document describes a retrospective review by a single expert to affirm the clinical value of the EEG records, not a comparative effectiveness study pitting human readers with vs. without AI assistance. The device is a physiological signal monitor, not an AI diagnostic tool in this context.

6. Standalone (Algorithm Only) Performance

• Standalone Performance Done? The "Software Verification Testing" section describes end-to-end testing of the system (acquiring, transmitting, and making data viewable), which is essentially testing the algorithm's ability to process and present the data. It verifies the functionality of the system components and data integrity. However, it's not a standalone diagnostic performance study (e.g., sensitivity/specificity for a given condition), as the device "does not make any diagnostic conclusion about the subject's condition." Its essential performance is to record and transfer EEG data.

7. Type of Ground Truth Used

• For Clinical Performance (Pediatric Extension): Expert consensus/review. The "experienced pediatric epileptologist" reviewed the REMI EEG data to assess its quality and presence of EEG features.
• For Technical Performance (Software, Electrical, etc.): Predetermined engineering specifications, compliance with recognized standards, and functional verification.

8. Sample Size for the Training Set

• The document does not mention a training set or machine learning model being the primary focus of this submission. The device is described as a "physiological signal monitor" that "does not make any diagnostic conclusion." The clinical experience discussed references "318 pediatric patients with a mean REMI Sensor wear of 1.7 days" from NIH-funded studies initially used for "REMI development efforts," which could have implicitly involved some level of data-driven development or refinement, but it's not explicitly framed as a "training set" for a distinct AI algorithm evaluated in this 510(k). The focus of this 510(k) is the extension of the device's indications to a younger pediatric age group and the new hydrocolloid sticker.

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

• As a training set is not explicitly referred to as part of a machine learning model specific to this 510(k)'s purpose, the method for establishing its ground truth is not detailed. The "REMI development efforts" data (n=318) was collected "under Institutional Review Board oversight." However, for the specific aspect of this 510(k) (pediatric extension), the ground truth for the test set (n=13) was established via retrospective review by an experienced pediatric epileptologist.

Ask a specific question about this device

The REMI Remote EEG Monitoring System is indicated for use in healthcare settings where near real-time and/or remote EEG is warranted and in ambulatory settings where remote EEG is warranted. REMI uses single patient, disposable, wearable sensors intended to amplify, capture, and wirelessly transmit a single channel of electrical activity of the brain for a duration up to 30 days.

The REMI System uses the REMI-Mobile software application that runs on qualified portable general purpose computing platforms. REMI-Mobile displays user setup information to trained medical professionals and provides notifications to medical professionals and ambulatory users. REMI-Mobile receives and transmits data from connected REMI Sensors to the secure REMI-Cloud where it is stored and prepared for review on qualified EEG viewing software.

REMI does not make any diagnostic conclusion about the subject's condition and is intended as a physiological signal monitor. The REMI System is indicated for use with adult and pediatric patients (6+ years).

The REMI System has three major components:

1. REMI Sensor A disposable EEG sensor which is placed on the patient's scalp using a conductive REMI Sticker
2. REMI Mobile A mobile medical application that is designed to run on a qualified commercial-off-the-shelf mobile computing platform (an Android tablet for use in healthcare settings, and a portable/wearable Android smartwatch for use in ambulatory settings), acquire EEG data transmitted from REMI Sensors and then transmit the EEG data and associated patient information via wireless encrypted transmission to,
3. REMI Cloud A HIPAA-compliant secure cloud storage and data processing platform where data is processed into a qualified EEG reviewing software format for neurological review.

This 510(k) submission includes the addition of the Android smartwatch for ambulatory use and increases the duration of monitoring to up to 30 days.

The provided text describes the REMI Remote EEG Monitoring System and its substantial equivalence to a predicate device. However, it does not include specific quantitative acceptance criteria or detailed study results that would typically be associated with performance metrics like sensitivity, specificity, accuracy, or effect sizes for AI assistance. The document focuses on demonstrating substantial equivalence through testing of electrical safety, wireless technology, software, and human factors.

Here's an attempt to answer your questions based on the available information, with acknowledgements where information is missing.

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

Based on the provided text, the acceptance criteria are generally framed around meeting regulatory standards and functional requirements rather than quantitative performance metrics for diagnostic accuracy.

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 does not specify sample sizes for any of the described tests. It mentions "testing conducted," "accelerated aging," "ship testing," and "human factors/usability testing," but provides no details on the number of units, subjects, or data points involved. Similarly, data provenance (country of origin, retrospective/prospective) is not mentioned.

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 document does not describe any study establishing ground truth with expert review for a diagnostic purpose. The device is explicitly stated to "not make any diagnostic conclusion" and is "intended as a physiological signal monitor." Therefore, this question is not applicable in the context of the provided information.

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

Since no expert-based ground truth establishment or diagnostic performance evaluation is detailed, there is no mention of an adjudication method.

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 MRMC study is mentioned. The device is a physiological signal monitor and does not involve AI assistance for human readers in a diagnostic context.

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

The device itself is a system for acquiring and transmitting EEG data for review by medical professionals on qualified EEG viewing software. It does not perform standalone diagnostic algorithms. Its "Essential Performance" is to record digitized EEG data and transfer it.

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

Given that the device is a physiological signal monitor and "does not make any diagnostic conclusion," the concept of "ground truth" as typically used for diagnostic or screening devices (e.g., pathology, expert consensus on a disease state) is not applicable here. The ground truth for its performance would be the accuracy of EEG signal acquisition and transmission, which is assessed through bench testing and compliance with electrical standards.

8. The sample size for the training set

The document does not describe any machine learning or AI-based component that would require a "training set." The software updates mentioned are for supporting new hardware (smartwatch) and extending monitoring duration, not for developing new diagnostic algorithms.

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

Not applicable, as no training set for an AI/ML algorithm is described.

Ask a specific question about this device