Search Results

MyPAO patient-specific guides are devices intended to be used as anatomical guides are specifically designed based on the CT images of the patient (i.e., pelvis, proximal femur). The guides are intended to assist the surgeon in the positioning of the acetabular fragment during periacetabular osteotomies to treat patients who require to undergo periacetabular osteotomy. MyPAO guides are intended for single use only.

The MyPAO Guides enlarge Medacta's MyPAO realignment system, already including MyPAO Planning report, in order to provide the surgeons a further help in the realignment phase after a Periacetabular Osteotomy.

Specifically, the current submission aims at obtaining the clearance for the changes made to MyPAO Planning report and for the following new devices:

• Reposition PSI guides (Pre-op drill guide, Ilium and Acetabulum Rep Guides, Realignment Arc):
• Preoperative and postoperative bone models;
• Reusable instruments to be used with the guides.

MyPAO Guides are patient specific guides designed on the basis of the patient's preoperative CT scan and subsequent preoperative planning and they aim at replicating the desired correction before the final fixation of the acetabular fragment.

The MyPAO Guides as well as their bone models are single-use and they are provided non-sterile.

The 3D pre-operative planning based on the patient's CT scans is performed through MyPAO software that is the same software already used to generate MyPAO Planning Report. The only change made to the already cleared MyPAO Planning report (K200589) regards the addition of two parameters in the report, i.e. 3D femoral head coverage and acetabular anteversion.

The provided text describes the MyPAO Guides, a medical device intended to assist surgeons during periacetabular osteotomies. While the document mentions performance data and clinical studies, it focuses on demonstrating substantial equivalence to a predicate device for regulatory clearance (510(k)), rather than establishing specific acceptance criteria for an AI/algorithm-driven device and proving its performance against those.

The MyPAO Guides are patient-specific anatomical guides based on CT images and preoperative planning, but the description does not indicate that the core "device" itself is an AI algorithm that makes diagnostic or treatment recommendations. The "MyPAO software" mentioned is for planning, and the guides themselves are physical devices.

Therefore, the information required to answer the prompt (acceptance criteria for an AI device, sample sizes for AI test sets, expert adjudication methods for AI ground truth, MRMC studies, standalone AI performance, etc.) is not present in the provided document, as it pertains to the clearance of physical surgical guides with software for planning.

However, I can extract the relevant performance data related to the clinical study cited for the MyPAO Guides, even though it's not structured as an AI acceptance criteria study.

Here's what can be extracted based on the provided text, interpreted in the context of the prompt's questions, acknowledging the limitations mentioned above:

Acceptance Criteria and Device Performance (for the clinical study performed related to MyPAO Guides, not an AI algorithm):

Details of the Clinical Study Related to MyPAO Guides:

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

   • Test Set Sample Size: 10 patients.
   • Data Provenance: The study was a "single-arm study with a literature control." It was "conducted outside of the United States." The data is prospective, involving actual patient treatments.

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

   • The "ground truth" for this study was the post-operative correction results compared to the pre-operative planned correction. Measurements were done using "2D planar x-rays to measure the final lateral center edge and acetabular index."
   • The document states "10 patients were treated by three surgeons using the MyPAO surgical planning software and guides." It doesn't specify if these 3 surgeons (or others) were involved in establishing the "ground truth" measurements from the x-rays, or their qualifications.

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

   • The document does not describe any specific adjudication method for the measurements.

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, an MRMC comparative effectiveness study was not done. The clinical study was a single-arm study evaluating the MyPAO system (guides + planning software) against a literature control for conventional surgical technique accuracy. This is not designed to assess human reader improvement with AI assistance.

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

   • The "MyPAO Guides" are physical devices. The "MyPAO software" is for planning. The study evaluated the system (software + guides) in facilitating surgical outcomes. No standalone algorithm performance study (where an AI algorithm performs measurements or diagnoses without human input) is described. The device's primary function is as an anatomical guide for human surgeons.

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

   • The ground truth was based on a comparison of pre-operative planned correction versus post-operative correction results measured from 2D planar x-rays. This is a form of outcomes data directly related to the surgical correction achieved. The acceptance criteria themselves were based on literature for conventional surgical accuracy, implying that the "ground truth" for acceptable performance was derived from expert consensus in the broader medical literature.

7. The sample size for the training set:

   • The document describes a 510(k) submission for a physical medical device (MyPAO Guides) assisted by planning software. It is not an AI algorithm clearance with explicit training and test sets in the typical AI/ML sense. Therefore, no "training set" sample size for an AI model is provided or applicable as per the document's content.

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

   • As there's no explicitly defined "training set" for an AI algorithm in this context, this question is not applicable to the provided document. The planning software's underlying data or algorithms are not detailed in terms of their "training" in this regulatory submission.

Ask a specific question about this device