Search Results

BWMini is an electroencephalograph, which is a device used to measure and record the electrical activity of the patient's brain obtained by placing two or more electrodes on the head.

BWMini is multi-channel (up to 32 channels) system designed for Electroencephalograph (EEG), Polysomnography (PSG) and Home Sleep Testing (HST) recording application, in research, home sleep studies, ambulatory and clinical environments.

The BWMini does not make any judgment of normality of the displayed signals or the results of an analysis. In no way are any of the functions represented as being in and of themselves diagnostic.

Device Description

BWMini is an electroencephalograph, which is a device used to measure and record the electrical activity of the patient's brain obtained by placing two or more electrodes on the head.

The BWMini does not make any judgment of normality of the displayed signals or the results of an analysis. In no way are any of the functions represented as being in and of themselves diagnostic.

The device is software based (Windows), uses a memory card for data storage, has a hard plastic external material, and is powered by batteries. It has up to 32 channels, 16 Bits AD Resolution, 50-60Hz Notch Filter, 1-500uv/mm Configurable Sensitivity Selection, 0.16 - 15Hz Configurable Low Frequency filters, 15 - 100 Hz Configurable High Frequency filters, and 2 DC Inputs. The user interface is IBM PC.

AI/ML Overview

The Neurovirtual BWMini EEG, HST, and PSG device is not an AI/ML powered device, therefore the typical acceptance criteria and study designs for such devices do not apply here. This document, K131335, is a 510(k) premarket notification for an electroencephalograph (EEG) device, which is a traditional medical device for measuring and recording electrical activity of the brain. The "acceptance criteria" and "study" described herein relate to the device's technical specifications and compliance with established medical device standards and performance testing, rather than AI model performance metrics like sensitivity, specificity, or AUC.

Here's a breakdown of the information provided, framed within the context of a traditional medical device submission:

1. Table of Acceptance Criteria and Reported Device Performance

For traditional medical devices like the BWMini, "acceptance criteria" refer to the specified technical performance parameters the device must meet, often derived from industry standards or clinical guidelines. "Reported device performance" then details how the device performed against these criteria in non-clinical testing.

Feature / Criterion	Acceptance Criteria (Target/Standard)	Reported Device Performance (BWMini)
Frequency Response	0.5Hz to 100Hz (acceptable frequency range for EEG based on "principal EEG authors").	Tested at: 0.5Hz, 1Hz, 2Hz, 3Hz, 4Hz, 5Hz, 10Hz, 15Hz, 20Hz, 25Hz, 30Hz, 40Hz, 50Hz, 70Hz, 80Hz, 90Hz, and 100Hz.
Accuracy of Frequency Response	Deviation of +/- 5% in the injected value at each tested frequency compared to a known and calibrated external source.	The conclusion states: "The BWMini meets the performance standards for perform EEG and PSG exams." This implies the +/- 5% deviation was met across all tested frequencies.
AD Resolution	Minimum resolution requirement for recording EEG according to ACNS Guideline.	16 Bits (exceeds the 12 Bits of one predicate and likely the ACNS guideline). (From Comparison Table)
Sensitivity Selection	Requirement for recording EEG according to ACNS Guideline.	1-500uv/mm Configurable (exceeds the range of predicate devices and likely the ACNS guideline). (From Comparison Table)
Low Frequency Filters	Requirement for recording EEG according to ACNS Guideline.	0.16 - 15Hz Configurable (exceeds the range of predicate devices and likely the ACNS guideline). (From Comparison Table)
Notch Filter	Standard for filtering interference.	50-60Hz (matches predicate devices). (From Comparison Table)
High Frequency Filters	Standard for filtering interference.	15 - 100 Hz Configurable (matches predicate devices). (From Comparison Table)
Compliance with Standards	IEC 60601-1, IEC 60601-1-1, IEC 60601-1-2, IEC 60601-2-26, IEC 60601-1-4, EN ISO 14971, EN ISO 13485, FDA Guidance software validation version 1.1	The device is stated to be "in compliance with the applicable clauses of the following standards." (From Section I: Safety and Effectiveness)

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

This is a non-clinical device performance test, not a clinical trial with patient data.

Sample Size for Test Set: Not applicable in the traditional sense of patient data. The "test set" here refers to the device units themselves that underwent performance testing. The document states "We test the produced units for performance..." implying testing was conducted on one or more manufactured units of the BWMini. The exact number of units or test runs is not specified but is typically part of a detailed test report (which is referenced as "ATTACHMENT 7").
Data Provenance (e.g., country of origin of the data, retrospective or prospective): Not applicable as this is hardware/software performance testing, not data collection from patients. The testing was conducted by Neurovirtual USA, Inc.

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

Not applicable for this type of device performance test. The "ground truth" for these tests comes from known, calibrated external signal sources as described in Section J: Non-clinical Testing. The acceptance criteria themselves are based on established guidelines (e.g., ACNS Guideline) and widely accepted principles from "principal EEG authors" rather than expert consensus on specific cases.

4. Adjudication Method (e.g., 2+1, 3+1, none) for the Test Set

Not applicable. Device performance testing against calibrated sources does not involve adjudication by multiple experts. The comparison is objective: device output vs. known input from a calibrated source, with a defined acceptable deviation.

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. This device is an acquisition system, not an AI-powered diagnostic tool that assists human readers. It records physiological signals. There is no AI component mentioned or implied in this 510(k) submission. Therefore, no MRMC study or effect size related to AI assistance would be conducted.

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

Not applicable. The BWMini is an electroencephalograph (EEG), polysomnograph (PSG), and home sleep testing (HST) recording application. It is a hardware and software system for data acquisition. The submission explicitly states: "The BWMini does not make any judgment of normality of the displayed signals or the results of an analysis. In no way are any of the functions represented as being in and of themselves diagnostic." This confirms it is not an algorithm that performs standalone diagnoses or interpretations.

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

For the non-clinical testing of the frequency response, the ground truth was data from a known and calibrated external source. This means a signal generator or similar equipment was used to inject precise, known frequencies and amplitudes into the device, and the device's recorded output was compared against these known inputs.

8. The Sample Size for the Training Set

Not applicable. This device is not an AI/ML device and therefore does not have a "training set" in the context of machine learning.

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

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

Ask a Question

Ask a specific question about this device

Page 1 of 1