Search Results

The Okti System is intended for use in the recording, displaying, analysis, printing and storage of human biological parameters such as heart and muscle activity, eye movement, breathing and body movements to assist in the diagnosis of various neurological disorders. The Okti is designed for use in a hospital or other clinical environment. The Okti is only to be used under the direction of a physician.

The Okti is a physiological data acquisition device intended for conducting EEG studies. It is used in conjunction with a PC running Compumedics Profusion EEG software. Power is provided by either an internal Lithium-Ion battery or a POE 100BASE-TX network connection. Physically, the Okti comprises a base unit housing the battery and digital electronic circuits. A changeable module connects to the base unit and houses the analogue circuity and electrode connectors providing options for 32, 64 or 128 referential, or single ended, channels primarily intended for EEG. Each module includes several differential inputs for electrode-based signals such as ECG, EMG or EOG. All these inputs are class CF for patient safety. Patient physiological data is digitised using multiple 24-bit ADCs and filtered using fixed point FIR filters implemented internally to an FPGA located in the analogue module. The data is transferred to a generalpurpose processor in the base unit over a synchronous serial link. From there the patient data may be saved to a device internal SD card or sent to a remote PC over an Ethernet or Wi-Fi LAN connection. A Bluetooth interface is also included for connection to external devices such as an oximeter. The Okti system is available in Okti32, Okti64, or Okti128. These differ only in channel number.

The provided text describes a medical device, Okti, which is an Electroencephalograph (EEG) system. The document is a 510(k) premarket notification to the FDA, asserting the substantial equivalence of Okti to a previously cleared predicate device, the Grael EEG system.

While the document extensively covers the electrical, mechanical, and performance specifications of the Okti device and compares it to its predicate, it does not describe a study involving AI assistance, human readers, or the establishment of ground truth by multiple experts for diagnostic performance evaluation, as would be expected for a typical AI/ML-driven device. The device described here is an EEG recording device, not an AI or diagnostic algorithm. Therefore, many of the requested criteria related to AI/ML device performance evaluation (e.g., sample size for test/training sets, data provenance, number of experts, adjudication, MRMC study, standalone performance, type of ground truth for diagnostic accuracy) are not applicable or not provided in this specific document.

The "performance data" summary relates to the technical validation of the device's ability to accurately record and process physiological signals, ensuring electrical safety, electromagnetic compatibility, and adherence to established EEG standards. It's about the quality of the signal acquisition, not the diagnostic accuracy of an AI interpreting the signals or a human interpreting with AI assistance.

Based on the provided text, here's what can be extracted regarding acceptance criteria and device performance, focusing on the device's technical performance rather than diagnostic AI performance:

Acceptance Criteria and Reported Device Performance (Technical Validation)

The document focuses on demonstrating substantial equivalence to a predicate device (Grael EEG) by comparing technical, electrical, and physical specifications, and by conducting standard performance and safety tests for EEG acquisition devices.

Table of Acceptance Criteria and Reported Device Performance (Technical Parameters):

Study Details (Technical Validation, Not Diagnostic Performance):

As the document pertains to the technical aspects of an EEG acquisition device, not an AI-driven diagnostic tool, the following points are addressed based on what is available:

1. Sample size used for the test set and the data provenance:

   • No "test set" in the context of diagnostic data (e.g., medical cases/images) is mentioned. The testing involves "extensive collection of tests" against electrical safety, EMC, and EEG specific performance standards.
   • No information on data provenance (country of origin, retrospective/prospective) because it's not a dataset for diagnostic evaluation.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Not applicable. The "ground truth" here is adherence to engineering and safety standards (e.g., a specific noise level is verifiable by instrumentation, not expert consensus on a diagnosis).

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

   • Not applicable. Adjudication is not relevant for technical performance testing against fixed standards.

4. 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. This is an EEG recording device, not an AI-assisted diagnostic tool.

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

   • No. This is an EEG recording device. Its "performance" is about accurate signal acquisition and processing, not about an algorithm making a diagnosis. The device's analysis capabilities are implicitly linked to its software, but it's not presented as an AI making diagnostic predictions.

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

   • The "ground truth" for this submission is based on engineering specifications, regulatory standards (IEC 60601-1, IEC 60601-1-2, IEC 80601-2-26), and direct comparison to a predicate device's established performance. It's about the device's physical and electrical characteristics meeting predefined benchmarks for an EEG system, not clinical diagnostic accuracy based on patient outcomes or expert readings.

7. The sample size for the training set:

   • Not applicable. This device is not an AI/ML diagnostic algorithm that requires a training set.

8. How the ground truth for the training set was established:

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

In summary: The provided document is for the regulatory clearance of a medical device (EEG hardware and associated software for acquisition) based on its technical specifications and substantial equivalence to a predicate, not for an AI/ML diagnostic algorithm whose performance would be assessed through clinical studies with expert-adjudicated ground truth.

Ask a specific question about this device