(256 days)
Use of the Artemis 123 Biomagnetometer is indicated for the patient whose physician believes that information about the magnetic fields produced by that patient's brain and information about the location of the sources of those magnetic fields could contribute to diagnosis or therapy planning.
The Tristan Technologies Artemis 123 Biomagnetometer (hereinafter referred to as the "Artemis 123") utilizes superconducting signal pickup coils and Superconducting Quantum Interference Devices (SQUIDs) to detect and amplify magnetic fields produced by electrical activity in brain. The Artemis 123 consists of a sensor unit, an electronics subsystem for preliminary amplification, filtering, and analog to digital conversion of the signals from each SQUID, an electronics rack containing power supplies to power the electronics subsystem, a computer to control the operation of the electronic subsystem and the SQUIDs and to acquire and store the signal values collected by the system, and a patient table which accommodates and facilitates the optimal positioning of the head of a human being adjacent to the sensor unit.
The Tristan Technologies Artemis 123 Biomagnetometer is indicated for use for the patient whose physician believes that information about the magnetic fields produced by that patient's brain and information about the location of the sources of those magnetic fields could contribute to diagnosis or therapy planning.
The study presented focuses on demonstrating technological equivalence to a predicate device, the Magnes 2500 WH Biomagnetometer System, rather than establishing specific clinical acceptance criteria based on a disease diagnosis or therapy planning outcome. Therefore, the "acceptance criteria" and "device performance" in the context of this submission relate to these technological equivalence metrics.
1. Table of Acceptance Criteria and Reported Device Performance
| Feature/Metric | Acceptance Criterion (Equivalent to Magnes 2500 WH) | Reported Device Performance (Artemis 123) |
|---|---|---|
| Sensitivity (Hospital Environment) | Noise level above 100 Hz well below 10 fT/√Hz (characteristic form of Magnes 2500) | Average noise showed the same characteristic form as that of the Magnes 2500, with the noise level above 100 Hz being well below the specification of 10 fT/√Hz. |
| Source Localization (Phantom) | Localization of a magnetic dipole source within 5 mm of the actual location (equivalent to Magnes 2500 WH performance) | Determination of the location of each dipole (from two dipolar sources) to within 5 mm of the actual location. This performance is also equivalent to the localization of dipoles in a phantom with the Magnes 2500 WH system. |
| Underlying Technology | Superconducting magnetometry | Superconducting magnetometry |
| Refrigeration Method | Solid conduction from liquid helium | Solid conduction from liquid helium |
| Data Flow | SQUID output digitized, stored on hard drive | SQUID output digitized, stored on hard drive |
| Indications for Use | Use for patients whose physician believes information about brain magnetic fields and their sources could contribute to diagnosis or therapy planning. | Use for patients whose physician believes information about brain magnetic fields and their sources could contribute to diagnosis or therapy planning. |
| Safety Standard | IEC-60601-1 | IEC 60601-1 |
| Average coil-to-coil spacing | 25 mm | 25 mm |
| Superconducting Amplifiers | dc SQUID | dc SQUID |
2. Sample Size for Test Set and Data Provenance
- Sample Size:
- Sensitivity Test: The noise spectra of all channels (123 pickup coils) were recorded from the Artemis 123.
- Source Localization Test: The magnetic field values were recorded for each of the Artemis 123 channels (123 pickup coils) from a phantom containing two dipolar sources.
- Data Provenance: Non-clinical tests conducted at the Children's Hospital of Philadelphia, USA. These tests were for "research use only" and the results were published in the peer-reviewed journal Frontiers Human Neuroscience on March 3, 2014. The data is retrospective in the context of this 510(k) submission, as it was published prior to the submission date.
3. Number of Experts and Qualifications for Ground Truth
This study did not involve human expert interpretation of brain magnetic field data for establishing ground truth. The "ground truth" for the non-clinical tests was established by:
- Sensitivity: The intrinsic noise characteristics of the device in an empty measurement environment, compared against the known specifications and characteristic noise form of the predicate device.
- Source Localization: The known actual locations of the two dipolar sources within the phantom. The "expertise" here lies in the precise engineering and construction of the phantom and the placement of the known sources.
4. Adjudication Method
Not applicable. This study did not involve human interpretation or subjective assessments that would require an adjudication method. The measurements were objective physical quantities (noise levels, localized source positions).
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No MRMC comparative effectiveness study was done. This submission focuses on the technological equivalence of the device's physical performance characteristics (sensitivity and source localization) compared to a predicate device, not on the impact of AI or the device on human reader performance in a clinical diagnostic setting.
6. Standalone Performance (Algorithm Only without Human-in-the-loop performance)
Yes, the performance evaluated was standalone performance of the device's physical sensing and localization capabilities.
- Sensitivity: The Artemis 123 was activated, and noise spectra were recorded and analyzed directly from its channels in an empty room.
- Source Localization: The Artemis 123 directly recorded magnetic fields from a phantom, and an algorithm fitted these values to a dipole source model to determine location, without human intervention in the interpretation process of the raw data for localization.
7. Type of Ground Truth Used
- Sensitivity: Based on the known specification (10 fT/√Hz) and characteristic noise profile of the predicate device (Magnes 2500 WH) as the comparative "ground truth."
- Source Localization: Physical ground truth established by the known and precise locations of dipolar sources within a specially constructed phantom.
8. Sample Size for the Training Set
No training set information is provided or relevant in this context. The study describes non-clinical performance evaluations of a physical device, not the training of an AI algorithm or machine learning model.
9. How Ground Truth for the Training Set was Established
Not applicable, as no training set was used.
§ 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).