Search Results

Blueprint® Mixed Reality system is indicated for use during Total Shoulder Arthroplasty using Stryker's FDA cleared implants that are also implants cleared for pre-operative planning with the Blueprint® Software. Blueprint® Mixed Reality system is intended to allow surgeons to visualize the Blueprint® 3D preoperative planning intra-operatively. Blueprint® Mixed Reality system is also indicated for stereotaxic surgery to guide the placement of a glenoid pin provided that registration between the patient's anatomical landmarks / surfaces can be established on the preoperative CT based plan. Blueprint® Mixed Reality system is to be only used for skeletally matured adult patients.

The Blueprint® Mixed Reality system is used to intraoperatively display stereoscopic threedimensional images of the Blueprint shoulder arthroplasty preoperative plan and to guide and enable the insertion of a pin following the reaming axis aimed at preparing the glenoid fossa surface to facilitate the placement of the implant. The Blueprint Mixed Reality system allows the user to both see the patient in real-time and their internal boney anatomy displayed on seethrough screens of a head-mounted device (goggles).

The Blueprint Mixed Reality system assists surgeons in visualizing stereoscopic threedimensional images of the patient's boney anatomy and intraoperatively aid by guiding the pin during placement through real-time feedback.

The Blueprint Mixed Reality system is composed of three main components including the HOLOBLUEPRINT software, Microsoft® HoloLens 2 Goggles, and Instrumentation.

The Blueprint Mixed Reality system runs the HOLOBLUEPRINT software application on the Microsoft HoloLens 2 using Microsoft Windows Holographic Operating System. The Blueprint Mixed Reality system leverages the reference device by receiving planned glenoid cases as input for visualizing and guiding the placement of the pin during the shoulder arthroplasty procedure. The digital components of the system interface with specialized surgical instrumentation to enable intra-operative guidance through real-time feedback to the surgeon.

The components of the subject device are as follows:

HOLOBLUEPRINT™ Software
HOLOBLUEPRINT is designed and written by Tornier SAS specifically for use on the Microsoft HoloLens 2 goggles (hardware). The HOLOBLUEPRINT software is to be installed on the hardware to be used by the surgeon as part of the Blueprint Mixed Reality system.

HoloLens 2 goggles (HoloLens 2)
The HoloLens 2 is a Microsoft hardware that runs Microsoft Windows 10 Holographic Operating System. HoloLens 2 is a see-through, mixed reality headmounted smart glasses.

Instrumentations (Instruments)
There are four reusable instruments and one single-use (sterile) instrument designed specifically for use with the Blueprint Mixed Reality system (subject device). The subject device system instruments include a Glenoid Pin Guide, Glenoid Digitizer, Instruments Check Block, Coracoid Clamp, and a Depth Stop Pin (Sterile).

The subject device is compatible with all Tornier (Stryker) commercially FDA-cleared glenoid implants (Except Patient Matched Implants) available in the reference device (K211359). The compatibility of implants with the reference device was validated and verification was performed for glenoid guidance and visualization of the patient's boney anatomy.

The provided document is a 510(k) premarket notification for the Blueprint Mixed Reality system. It describes the device's indications for use, technological characteristics, and non-clinical performance evidence to demonstrate substantial equivalence to a predicate device.

Based on the content, here's a description of the acceptance criteria and the study that proves the device meets them:

Acceptance Criteria and Reported Device Performance

The acceptance criteria are primarily focused on the accuracy and repeatability of the system, and that the software performs as per requirements and specifications. The document states that the system was validated through cadaver testing to demonstrate its accuracy and establish performance equivalency to the predicate device.

Study Details:

1. Sample Size and Data Provenance:

   • Test Set Sample Size: The document mentions "cadaver testing" and "simulated use cases" for validation. It does not specify the numerical sample size (e.g., number of cadavers or simulated cases) used for the test set.
   • Data Provenance: The document does not explicitly state the country of origin of the data. The company is based in France, but the testing locations are not specified. The studies were non-clinical, involving cadavers and simulated use. It is a prospective study in the sense that the testing was performed specifically to evaluate this device.

2. Number of Experts and Qualifications:

   • The document does not provide details on the number of experts used or their specific qualifications (e.g., "radiologist with 10 years of experience") for establishing ground truth or evaluating performance. The testing involved "cadaver testing" and "simulated use cases," implying technical personnel or surgeons would be involved in assessing accuracy, but their roles and expertise are not elaborated upon.

3. Adjudication Method for the Test Set:

   • The document does not specify an adjudication method (e.g., 2+1, 3+1) for the test set. Given it's a technical accuracy and performance study on a medical device system, it's more likely that technical measurements and predefined tolerance limits were used for "ground truth" rather than subjective expert consensus requiring adjudication.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

   • No, an MRMC comparative effectiveness study was not explicitly done. The document states, "Clinical testing was not deemed necessary for the determination of substantial equivalence." The evaluation was based on non-clinical performance evidence (bench testing, system accuracy validation through cadaver testing, functional and performance testing, software verification/validation). Therefore, there is no reported effect size of how much human readers improve with AI vs. without AI assistance.

5. Standalone (Algorithm Only) Performance:

   • Yes, a form of standalone performance was assessed. The "Bench Testing conducted to demonstrate the HOLOBLUEPRINT™ software performs as per the systems requirements and specifications" can be considered an assessment of the algorithm's performance in isolation from the full human-in-the-loop use case, focusing on its computational accuracy and adherence to specifications. The "System Accuracy Validation... through cadaver testing" also evaluates the system's (including the software) ability to provide accurate guidance measurements, independent of a specific surgeon's subjective interpretation (though a surgeon would utilize the output).

6. Type of Ground Truth Used:

   • The ground truth for the accuracy studies (cadaver testing) would likely be established through highly precise measurements using a gold-standard reference system (e.g., coordinate measuring machine, highly accurate optical tracking system, or direct physical measurements) to verify the device's reported position and orientation accuracy against the true anatomical features. For software performance, the ground truth would be defined by the pre-defined system requirements and specifications.

7. Training Set Sample Size:

   • This document describes the validation of a medical device system, not specifically a machine learning model where a distinct "training set" for an AI algorithm would be relevant. While the system incorporates "HOLOBLUEPRINT™ Software," the text does not indicate that this software is an adaptive AI model requiring a separate training phase with a specific data set size. Therefore, no sample size for a training set is provided or implied.

8. How Ground Truth for Training Set was Established:

   • Given that the document does not describe the use of an adaptive AI model with a training set, the concept of establishing ground truth for a training set is not applicable here. The software validation is against pre-defined specifications and requirements, not against an annotated training dataset for learning.

Ask a specific question about this device