(33 days)
The StealthStation™ System, with StealthStation™ Cranial Software, is intended as an aid for locating anatomical structures in either open or percutaneous neurosurgical procedures. Their use is indicated for any medical condition in which the use of stereotactic surgery may be appropriate, and where reference to a rigid anatomical structure, such as the skull, can be identified relative to images of the anatomy.
This can include, but is not limited to, the following cranial procedures (including stereotactic frame-based and stereotactic frame alternatives-based procedures):
- · Tumor resections
- General ventricular catheter placement
- · Pediatric ventricular catheter placement
- · Depth electrode, lead, and probe placement
- · Cranial biopsies
The StealthStation™ Cranial Software v1.3.0 works in conjunction with an Image Guided System (IGS) which consists of clinical software, surgical instruments, a referencing system and platform/computer hardware. Image guidance, also called navigation, tracks the position of instruments in relation to the surgical anatomy and identifies this position on diagnostic or intraoperative images of the patient. During surgery, positions of specialized surgical instruments are continuously updated on these images either by optical tracking or electromagnetic tracking.
Cranial software functionality is described in terms of its feature sets which are categorized as imaging modalities, registration, planning, interfaces with medical devices, and views. Feature sets include functionality that contributes to clinical decision making and are necessary to achieve system performance.
The changes to the currently cleared StealthStation S8 Cranial Software are as follows:
- . Addition of an optional image display that allows the user to see through outer layers to increase the visibility of other models.
- . Update the imaging protocol to support overlapping slices.
- . Minor changes to the software were made to address user preferences and to fix minor anomalies.
The provided document is a 510(k) premarket notification summary for Medtronic's StealthStation Cranial Software v1.3.0. It describes the device, its intended use, and a comparison to a predicate device, along with performance testing.
Here's an analysis to address your specific questions:
1. A table of acceptance criteria and the reported device performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| System Accuracy Requirements | |
| 3D Positional Accuracy: mean error ≤ 2.0 mm | Mean error ≤ 2.0 mm |
| Trajectory Angle Accuracy: mean error ≤ 2.0 degrees | Mean error ≤ 2.0 degrees |
2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)
The document does not specify a sample size (e.g., number of cases or images) for the performance testing. It states that the performance was determined "using anatomically representative phantoms and utilizing a subset of system components and features that represent the worst-case combinations of all potential system components."
Regarding data provenance, the testing was conducted in "laboratory and simulated use settings" using "anatomically representative phantoms." This indicates that the data was generated specifically for testing purposes, likely in a controlled environment, rather than being derived from real patient scans. The country of origin for the data is not specified, but the applicant company, Medtronic Navigation Inc., is based in Louisville, Colorado, USA. The testing appears to be prospective in nature, as it was specifically carried out to demonstrate equivalence.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience)
The document does not mention the involvement of human experts for establishing ground truth for the performance testing. The accuracy measurements (3D positional and trajectory angle) are typically derived from physical measurements against known ground truth (e.g., phantom dimensions, known instrument positions) in the context of navigation systems, not by expert consensus on image interpretation.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set
Not applicable. The performance testing described is objective measurement against physical phantoms, not subjective assessment by experts requiring adjudication.
5. 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. The document explicitly states: "Clinical testing was not considered necessary prior to release as this is not new technology." This device is an image-guided surgery system software, not an AI-assisted diagnostic tool that would typically undergo MRMC studies. The changes in this version (v1.3.0) are described as "minor changes to the software were made to address user preferences and to fix minor anomalies" and "Addition of an optional image display that allows the user to see through outer layers," suggesting incremental updates rather than a fundamentally new AI algorithm.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done
Yes, the performance testing was effectively "standalone" in the sense that the system's accuracy was measured against a known physical ground truth (phantoms) rather than evaluating human performance with the system. The reported accuracy metrics describe the device's inherent precision in tracking and navigation, independent of user interaction during the measurement process itself.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
The ground truth used was based on known physical properties of anatomically representative phantoms. This means that a physical phantom with precisely known dimensions and features was used, and the device's ability to accurately locate points and trajectories within that known physical structure was measured. This is a common and appropriate method for validating the accuracy of surgical navigation systems.
8. The sample size for the training set
Not applicable. This device, as described, is a software for image-guided surgery, not an AI/ML model that would typically have a "training set" in the context of deep learning. The changes are described as minor software updates and an optional display feature, not a new algorithm requiring a training phase from data.
9. How the ground truth for the training set was established
Not applicable, as there is no mention of a training set for an AI/ML model in this submission.
§ 882.4560 Stereotaxic instrument.
(a)
Identification. A stereotaxic instrument is a device consisting of a rigid frame with a calibrated guide mechanism for precisely positioning probes or other devices within a patient's brain, spinal cord, or other part of the nervous system.(b)
Classification. Class II (performance standards).