(226 days)
The Orion LifeSpan™ MEG non-invasively measures the magnetoencephalographic (MEG) (and, optionally, electroencephalographic (EEG) signals) produced by electrically active tissue of the brain. These signals are recorded by a computerized data acquisition system, displayed, and may then be interpreted by trained physicians to help localize these active areas. The locations may then be correlated with anatomical information of the brain. MEG is routinely used to identify the locations of visual, auditory, and somatosensory in the brain when used in conjunction with evoked response averaging devices. MEG is also used to non-invasively locate regions of epileptic activity within the brain. The localization information provided by the device may be used, in conjunction with other diagnostic data, as an aid in neurosurgical planning.
It is assumed that the users of the Orion LifeSpan™ MEG are physicians or neurology laboratory technicians who have received training in the following areas:
• Hospital procedures
• Physiological monitoring of patients
• Training relevant to the specific discipline or disorder under investigation
Note: This indication for use specifically excludes use of the Orion LifeSpan™ MEG as life support equipment, for example vital signs monitoring in intensive care units.
The Orion LifeSpan™ MEG is a magnetoencephalograph (MEG) which records magnetic signals from the human brain. The Orion LifeSpan™ MEG uses the CURRY software platform (K001781) to acquire, process and display these signals. Optionally EEG can be recorded simultaneously with the MEG using an integrated SynAmps2 (K023771) hardware system.
The Orion LifeSpan™ MEG can optionally be provided with adult, child or both sized helmets.
The Orion LifeSpan™ MEG is used by skilled operators or physicians trained in the acquisition and interpretation of such signals. It is used as an adjunct as part of a range of measurements performed, such as imaging and other studies, to form a more complete picture of the patient's pathology alongside a standard clinical workup.
Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Device Name: Orion LifeSpan™ MEG
Predicate Device: Elekta Neuromag
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Empty-room noise performance: Noise level not greater than the predicate (Elekta Neuromag) | Measured average noise performance was comparable to Elekta Neuromag (Orion: 10.54 fTrms/√Hz; Elekta: 11.31fTrms/√Hz). |
| Phantom Localization Accuracy: Average localization error and maximum localization error less than one standard deviation away from the predicate (0.7mm total error across all 3 dimensions) | Average localization error difference was 0.4mm (Orion: 2.3mm; Elekta: 1.9mm). Maximum localization error difference was 0.5mm (Orion: 4.22mm; Elekta: 3.72mm). Both are less than one standard deviation away (0.7mm). |
| Device Compatibility: Addition of compatible devices (HPI Coils, EEG System, evoked response stimulators) should not significantly increase the noise level and not exceed the noise threshold (10fTrms/√Hz at around 100Hz) | The empty room test results showed that the addition of each compatible device did not significantly increase the noise level of the Orion LifeSpan™ MEG and did not exceed the noise threshold. |
| CURRY Software Verification and Validation: All relevant CURRY Software specifications verified and validated in accordance with IEC 62304:2006. | CURRY Software verification and validation tests show that the CURRY Software supports the Orion device functionalities in a manner comparable to the predicate (i.e., processing, display, localization, MEG control functions). |
| Limited Channel Operation: Average localization error below 2mm, and maximum localization error below 4mm, when operating at 80% full channel capacity. | Average localization error range: 0.05-0.09mm. Maximum localization error range: 0.14-0.17mm when operating without 20% of its channels (80% full channel capacity). These values are well within the specified limits. |
2. Sample Size Used for the Test Set and Data Provenance
The document does not explicitly state a "sample size" in terms of number of patients or cases for the non-clinical performance studies. Instead, it describes:
- Empty-room noise performance & Device compatibility tests: Performed using the Orion LifeSpan™ MEG and the predicate Elekta Neuromag devices themselves in an empty, magnetically shielded room. The data provenance includes a specific measurement of the Elekta Neuromag TRIUX MEG at Swinburne University of Technology, dated November 6th, 2019. This indicates retrospective use of the predicate device's data for comparison.
- Phantom Comparison Test: An "identical phantom signal" was recorded with both systems. This implies one or more phantom measurements. Data provenance again includes the Elekta Neuromag TRIUX MEG measured at Swinburne University of Technology.
- Limited Channel Operation Test: Phantom localization testing was performed. The number of phantom trials or repetitions for this test is not specified, but it was performed at three different degrees of channel operation (95%, 90%, and 80% of full channel capacity).
There is no mention of human subject data (clinical data) in these performance studies.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of those Experts
Not applicable. The ground truth for the non-clinical tests (empty-room noise, phantom localization, device compatibility, software verification) was established by physical measurements and engineering specifications, not by human expert interpretation of medical images or signals.
4. Adjudication Method for the Test Set
Not applicable, as no human expert review or adjudication process was described for establishing ground truth in these non-clinical tests.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done
No, an MRMC comparative effectiveness study was not done. The document explicitly states: "Clinical testing was not performed." The studies presented are non-clinical, comparing device performance metrics directly to a predicate device or pre-defined engineering criteria, not evaluating human reader performance with or without AI assistance.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done
The device, the Orion LifeSpan™ MEG, is a medical instrument (magnetoencephalograph/electroencephalograph with associated software) for measuring brain signals, not an AI algorithm. Therefore, the concept of a "standalone algorithm performance" study as typically understood for AI/ML devices is not directly applicable. The non-clinical studies evaluate the standalone performance of the device itself (noise levels, localization accuracy) against established benchmarks. The CURRY software, while a component, is verified for its functionalities like processing, display, and localization, which are intrinsic to the device's operation.
7. The Type of Ground Truth Used
The ground truth for the non-clinical studies was based on:
- Physical measurements and established engineering benchmarks: For noise levels (fTrms/√Hz).
- Known phantom signal locations: For localization accuracy, where the true source location in the phantom is precisely known.
- Compliance with international standards: IEC 60601-1, IEC 60601-1-2, IEC 80601-2-26, IEC 62304, ISO 14971.
8. The Sample Size for the Training Set
Not applicable. This device is a measurement instrument, not an AI/ML algorithm that requires a "training set" in the context of machine learning. The software component, CURRY, undergoes verification and validation but not "training" in the AI sense.
9. How the Ground Truth for the Training Set was Established
Not applicable, as there is no training set for an AI/ML algorithm described for this device.
§ 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).