FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview

Intended Use

Neuro Omega System:

The Neuro Omega System with the incorporated HaGuide Software, including the Drive Headstage unit, is intended to assist neurosurgeons in the operating room during functional neurosurgery and to record from and stimulate brain motor and sensory neurons and to aid in the placement of depth electrodes.

The Neuro Omega System with the incorporated HaGuide Software is also intended:

To monitor, record, and display the bioelectric signals produced by muscles, to stimulate peripheral nerves, and to monitor, record, and display the electrical activity produced by nerves to aid the clinician in the diagnosis and prognosis of neuromuscular disease (EMG).
To measure, record, and display the electrical activity of the patient's brain obtained from two or more electrodes on the head (EEG).
To measure, display and record the electrical activity of the patient's brain obtained from ECOG strip and grid electrodes.
To provide stimulation via electrode pairs or a hand-held bipolar probe for use in functional brain mapping procedures during treatment of patients with a seizure disorder.

The Neuro Omega System with the incorporated installed HaGuide Software, is intended for intraoperative use by medical personnel. Within hospitals, laboratories, clinics, or nursing home settings or outside of a medical facility under the direct supervision of a medical professional. The device may also be placed in the intensive care unit or operating room for continuous recording.

NeuroSmart System:

The NeuroSmart System with the incorporated HaGuide Software, is intended to be used in assisting neurosurgeons, in the operating room during functional neurosurgery, to record from and stimulate brain motor and to aid in the placement of depth electrodes

Device Description

The cleared devices under K220553, the HaGuide Software incorporated in Neuro Omega/NeuroSmart systems, is designed to detect the STN region, it detects the entrance and exit boundaries of STN regions. Furthermore, it presents real-time graphs of power spectrum density and the normalized root mean square (nRMS) of the region. The HaGuide Software consists of the following components:

Patient Dashboard: The Patient Dashboard is comprised of a list of patients/cases that the user added, providing a user-friendly way to manage the surgery cases.
Planning Setup: The Planning Setup is used by the user to setup the surgery planning.
Trajectory Setup: The Trajectory Setup is used by the user to setup trajectory.
MER/HaGuide: MER/HaGuide is comprised of:
- Real-time Microelectrode Recording (MER) signal processing, presented as o Normalized Root Mean Square (NRMS) and Power Spectrum Density (PSD) graphs, to assist the user to interpret the signals.
- MER notes taking adding notes and tags to specific site location to o document and annotate the signals.
- HaGuide Recommendation: Detection of the Subthalamic Nucleus (STN) o borders, including the Intra STN detection of Dorsolateral Oscillatory Region (DLOR) and Ventromedial Non-Oscillatory Region (VMNR) boundary.
Operation Room Report Upload – Cloud: The OR report uploading feature is used to export the case data and HaGuide graphs and HaGuide recommendation securely to the cloud.
Raw Data Upload – Cloud: The raw data uploading feature is used to upload the raw recording data files securely to the cloud for offline processing

The HaGuide Software can work in three setups:

1. The HaGuide Software can be installed on the Neuro Omega/NeuroSmart PC and used on the system.
1. The HaGuide Software can be installed on an external PC as a standalone software for the purpose of working in off-line mode with no connection to the system.

In all setups the safety and effectiveness of the Neuro Omega/NeuroSmart systems isn't compromised, as the HaGuide Software is a non-blocking software (it doesn't affect the system's functionality even in malfunction). The HaGuide Software can run on both modes without affecting the Neuro Omega/NeuroSmart systems safety and effectiveness, nor its functionality.

AI/ML Overview

The provided text describes the regulatory clearance for the Neuro Omega System and NeuroSmart System, which incorporate the HaGuide Software. It focuses heavily on the software as a standalone component and its integration with the existing systems.

Based on the provided text, the study proving the device meets acceptance criteria is primarily non-clinical performance testing of the HaGuide Software and its integration with the Neuro Omega/NeuroSmart systems, rather than a clinical study evaluating human-in-the-loop performance or diagnostic accuracy with ground truth from clinical data.

Here's an analysis of the requested information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Test Performed	Acceptance Criteria	Reported Device Performance
Software / System Verification	The acceptance criteria were defined according to the use cases and regressions tests performed by the Neuro Omega / NeuroSmart systems, to ensure proper functionality:

HaGuide SW integration with Alpha Omega cloud is tested.
The verification tests cover the design inputs (SRS) as well as use cases and sanity tests. | All tests passed the acceptance criteria. |
| Penetration Testing | The identified risks are determined, then the scales are passed in accordance to NIST SP 800-30r1. | Overall cybersecurity risks related to Confidentiality, Integrity, and Availability have been evaluated. This includes risks related to software supply chain. Any risks introduced by risk controls were evaluated and mitigated to an acceptable level. |
| EMC Reports (IEC 60601-1-2) | The subject devices are safe and effective and meet the requirements of IEC 60601-1-2:2014, IEC 60601-1-2:014/AMD1:2020. | All tests passed the acceptance criteria. |
| Safety Reports (IEC 60601-1) | Neuro Omega/NeuroSmart systems incorporated the HaGuide Software are safe and effective and meet the requirements of IEC 60601-1:2005, IEC 60601-1:2005/AMD1:2012, IEC 60601-1:2005/AMD2:2020. | All tests passed the acceptance criteria. |

Note: The provided document describes non-clinical performance tests focused on software verification, cybersecurity, electromagnetic compatibility (EMC), and electrical safety. It does not contain information about clinical performance such as diagnostic accuracy, sensitivity, or specificity in a medical context. The "HaGuide Recommendation" feature for STN border detection is mentioned, but no specific performance metrics or clinical study results for this feature are provided in the excerpt. The submission is framed as a 510(k) for a "new infrastructure and GUI" of the HaGuide software, maintaining the same purpose as the predicate, which implies a focus on demonstrating that the changes do not introduce new safety or effectiveness issues, rather than proving new clinical effectiveness.

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

Test Set Sample Size: Not explicitly stated as a "sample size" in the context of clinical data. The tests performed are engineering-focused:
- Software/System Verification: This involves internal regression tests and testing against "use cases and sanity tests." The number of test cases or specific data points tested is not enumerated.
- Penetration Testing: This involves analyzing and identifying risks, not a numerical "test set" of patient data.
- EMC/Safety Reports: These are engineering compliance tests performed on the physical devices.
Data Provenance: Not applicable in the context of clinical data for these non-clinical performance tests. The document does not describe the use of patient data for testing device performance related to medical diagnosis or treatment. The HaGuide Software's ability to "detect the STN region" and present "real-time graphs of power spectrum density and the normalized root mean square (nRMS)" suggests it processes physiological signals, but how these processed signals were specifically "tested" for accuracy against a clinical ground truth is not detailed in this section. The mention of "Raw Data Upload – Cloud" for "offline processing" suggests data collection, but not its use as a test set for a performance study.

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

Not applicable. The provided document describes non-clinical performance tests, not a clinical study involving human experts establishing ground truth for patient data.

4. Adjudication Method for the Test Set

Not applicable. There is no mention of a clinical test set requiring adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No. The document does not describe an MRMC study or any clinical comparative effectiveness study involving human readers with and without AI assistance. The focus is on the functional verification and safety of the software and system.

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

The HaGuide Software is described as having a "standalone software" mode for "off-line processing" and for "visual purpose only, meaning receive data from the systems and present it in graphs." While it can run in standalone mode, the "performance" described in the acceptance criteria table is for functional verification, cybersecurity, and electrical safety, not for a standalone clinical performance evaluation of its algorithms (e.g., accuracy of STN detection). The document states, "No changes are made to the retrieved data. Furthermore, the data will be used for verification only for the neurologist prior to fine-tuning the IPG work mode by using the IPG interface (not Alpha Omega's device)." This suggests the software provides information to clinicians, but its algorithmic accuracy in providing medical insights is not the subject of the detailed performance tests presented.

7. The Type of Ground Truth Used

For the reported non-clinical performance tests:

Software/System Verification: Ground truth is established by the defined "design inputs (SRS) as well as use cases and sanity tests." This is based on functional and technical specifications, not clinical outcomes or expert consensus.
Penetration Testing: Ground truth is against cybersecurity best practices and standards (e.g., NIST SP 800-30r1).
EMC/Safety Reports: Ground truth is against established international standards (IEC 60601-1-2 and IEC 60601-1).

No clinical ground truth (e.g., pathology, surgical findings, long-term outcomes, or expert consensus on clinical images/signals) is mentioned as being used for the performance evaluation described.

8. The Sample Size for the Training Set

Not applicable. The document does not describe the use of machine learning or deep learning models that would require a "training set" for the HaGuide Software's core functionality. While the software "detects the STN region," it's not specified if this detection relies on a trained AI model or deterministic algorithms. The "new infrastructure and GUI" suggests software updates rather than new AI model development requiring a training set.

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

Not applicable, as no training set for an AI model is described.