(66 days)
Grass Brainwave Software is intended for use by qualified research and clinical professionals with specialized training in the use of EEG/PSG recording instrumentation for the digital recording, playback, and analysis of physiological signals. Its specifications and features make it especially well suited to electroencephalography and polygraphic sleep recordings.
The Grass® Night Vision is a Microsoft Windows95™ application software package for monitoring, recording, and reviewing physiological signals. It includes features for assisting the user with making measurements, marking significant events, and tabulating the events for easier interpretation. The software consists of two major modules, the Recorder and the Reviewer, and several minor modules for setup, montage editing, file management, and report generation. Together with a personal computer and a set of physiological signal conditioners (like the Grass® Model 15 amplifier system), the plight Vision software enables the user to replace or upgrade traditional strip-chart recorders used in the sleep laboratory with modern, digital recording and review methods.
This document is a 510(k) summary for the Grass® NightVision Software, which is described as a Polysomnograph/Electroencephalograph Software. The information provided focuses on the device's intended use, functionality, and comparison to legally marketed predicate devices.
Crucially, this document does not contain a study proving the device meets specific acceptance criteria in the typical sense of a performance study with quantitative outcomes.
Instead, the document primarily makes a claim of "substantial equivalence" to existing devices based on similar design, function, and intended use. The "acceptance criteria" presented below are inferred from the device's description and its stated functionality, as no explicit performance metrics are provided in the document for statistical analysis.
Based on the provided information, here's a breakdown:
1. Table of Acceptance Criteria and Reported Device Performance
Given that this is a 510(k) submission primarily establishing equivalence and describing functionality for a software product in 1997, explicit "acceptance criteria" with quantifiable performance metrics (like sensitivity, specificity, accuracy) and a "reported device performance" section with statistical results from a study are not present in this document.
The "acceptance criteria" can be broadly interpreted as the software's ability to successfully perform its described functions as a digital polysomnograph/electroencephalograph software. The "reported device performance" in this context refers to the described features and capabilities of the software.
| Acceptance Criteria (Inferred from functionality) | Reported Device Performance (as described) |
|---|---|
| Real-time monitoring and display of physiological signals. This implies accurate and timely graphical representation of signals. | The Recorder module sequenially samples, digitizes, and displays up to 32 channels on-screen "in a real-time graphical 'chart' view," mimicking traditional strip-chart recorders. User can vary settings (color, position, sensitivity) and make event marks. Display parameters include trace on/off, color, line style, grid width, baseline, sensitivity, digital filters, and digital display on/off. Vertical resolution is 12-bits. Recorder sample rate: 0.1 - 1000Hz, all channels. |
| Digitization and storage of signals to computer's hard drive. This implies reliable data capture and saving. | Signals are digitized and stored to the computer's hard drive, creating a waveform data file and a log file of annotations. |
| Replay of recorded data on-screen. This implies accurate retrieval and display of stored signals for review. | The Reviewer module allows technologists/physicians to review previously recorded data files on-screen with controls for paging, time scale adjustment, trace settings, and making measurements. |
| Assistance with marking and tabulation of diagnostically significant events and measurements. This implies user-friendly tools for analysis. | The Reviewer module provides tools for tagging each "page" with a "sleep stage" and adding event marks. Cursor measurements (amplitude, time, duration, min, max, std.dev., integral, area, rate, peak-peak) are available. Analysis Functions & scoring aids include trend, zoom, FFT, hypnogram, goto mark, goto epoch, goto time, marker info. Real-time annotation allows 23 predefined markers and one comment key. |
| Generation of summary reports and graphs for technologist/physician review. This implies creation of digestible summaries from raw data. | The Report Generator module processes the "scoring file" into a readable, condensed, and formatted report, creating totals, averages, minimums, maximums, and correlating events. Report includes: Sleep summary; arousal summary; latencies; respiratory disturbance summaries, details and indices; O2sat analysis; correlations; graphical displays of selected channels, trends, events, and hypnograms. |
| Ability to be used by qualified research and clinical professionals for digital recording, playback, and analysis of physiological signals, especially EEG and polygraphic sleep recordings. This is the overall intended use. | The software is explicitly stated as intended for this use, replacing or upgrading traditional strip-chart recorders. It supports 32 channels, multiple recording montages, and offers computer-assisted (non-automated) scoring. |
| Non-diagnostic nature. The software should not automatically make diagnostic decisions or interpret clinical significance. | Explicitly stated: "The Grass® Night Vision software relies completely on manual user assignments of sleep stage and other event marks." "Unlike some competitive products on the market, it makes no determination of clinically significant events automatically using algorithms (like apneas or sleep stages)." "The software does not make any judgement of normality or abnormality of any displayed or tabulated events or any analysis. The software is not represented as being in and of itself diagnostic." |
| Compliance with expert guidelines for manual review. | Grass® Instruments supports the position of the American Sleep Disorders Association that every page of the recording must be manually reviewed. The software aims for "faster scoring - without sacrificing the quality and thoroughness that hands-on, expert scoring guarantees." |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size for Test Set: Not applicable / Not mentioned. The document does not describe a clinical performance study using a "test set" of patient data in the context of evaluating diagnostic accuracy or a specific performance metric. The submission focuses on device functionality and equivalence.
- Data Provenance: Not applicable. Without a defined test set, there is no mention of data provenance (e.g., country of origin, retrospective/prospective).
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
- Number of Experts: Not applicable / Not mentioned. As no test set for performance evaluation is described, there's no mention of experts establishing a ground truth for such a set.
- Qualifications of Experts: Not applicable. The document does mention that the software is for "qualified research and clinical professionals with specialized training in the use of EEG/PSG recording instrumentation," and refers to the American Sleep Disorders Association guidelines for manual review by experts. However, these are general statements about the intended user and standard practice, not specific to a performance study's ground truth establishment.
4. Adjudication Method for the Test Set
- Adjudication Method: Not applicable / Not mentioned. No test set or performance study requiring an adjudication method is described.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- MRMC Study Done?: No. There is no mention of an MRMC comparative effectiveness study in this document. The software explicitly states it does not use automated algorithms for staging sleep or scoring respiratory events; it merely assists human users. Therefore, a study comparing human readers with AI assistance versus without AI assistance would not be relevant for this particular device's described function.
- Effect Size of Human Readers Improve with AI vs. without AI assistance: Not applicable.
6. Standalone (Algorithm Only) Performance Study
- Standalone Study Done?: No. The document explicitly states: "Night Vision uses the power of the computer whenever possible to aid and speed the scoring and review of the data files by keeping track of relevant signal information... Unlike some competitive products on the market, it makes no determination of clinically significant events automatically using algorithms (like apneas or sleep stages)." Therefore, a standalone performance study of an algorithm's diagnostic capabilities would not be relevant, as the device doesn't perform such automated diagnostics. Its role is as a tool for manual review and data presentation.
7. Type of Ground Truth Used
- Type of Ground Truth: Not applicable / Not mentioned for performance evaluation. Given the device's role as an assistant for manual human scoring and review, the "ground truth" ultimately relies on the interpretation of the "qualified research and clinical professionals." The software itself does not generate a diagnostic output or classification that would require an independent ground truth for validation in the context of a performance study. For the training set, see point 9.
8. Sample Size for the Training Set
- Sample Size for Training Set: Not applicable / Not mentioned. As the software "makes no determination of clinically significant events automatically using algorithms," it does not employ machine learning algorithms in a way that would necessitate a "training set" of annotated data for an AI model. The software's functionality is based on traditional software programming logic to display, record, and organize data according to user inputs and specifications.
9. How the Ground Truth for the Training Set was Established
- How Ground Truth for Training Set was Established: Not applicable. Since there is no indication of machine learning or an AI algorithm that requires a "training set" with ground truth, this question is not relevant to the described device. The software's "ground truth" during its development would be its adherence to established specifications for data acquisition, display, storage, and reporting, and its ability to accurately reflect user-entered information and perform calculations as programmed (e.g., cursor measurements, report generation based on entered scores).
§ 882.1400 Electroencephalograph.
(a)
Identification. An electroencephalograph 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.(b)
Classification. Class II (performance standards).