(85 days)
The DePuy Mitek MILAGRO BR Interference Screws are designed to attach soft tissues to bone in orthopedic surgical procedures for following indications: Shoulder: Proximal Biceps Tenodesis, Acromio-Clavicular Repair Elbow: Distal Biceps Tenodesis, Ulnar Collateral Ligament repair Knee: Collateral Ligament Repair
The DePuy Mitek Milagro® Interference Screws are absorbable, tapered, cannulated, threaded fasteners for use in interference fixation of soft tissue grafts or bone-tendon grafts. The Interference Screw is made from a composite made of absorbable Poly (lactide-co-glycolide) polymer and Tricalcium Phosphate (TCP). Modular Drivers, Ratchet Handle, Guidewires, a Tap and Notcher are provided separately as reusable accessories to assist in the placement of the MILAGRO BR Screw. The proposed DePuy Mitek Milagro® Interference Screws are offered in diameters of 5, 6mm and in the length of 12mm.
The provided text describes the DePuy Mitek Milagro® Interference Screws and their substantial equivalence to predicate devices, but it does not contain information about acceptance criteria or a study designed to prove the device meets specific acceptance criteria in the context of an AI/ML model for medical diagnosis or prognosis.
The document is a 510(k) summary for a medical device (interference screws) seeking clearance from the FDA. The "acceptance criteria" and "study" described in the prompt refer to the typical evaluation process for AI/ML-driven medical devices, often involving performance metrics like sensitivity, specificity, AUC, and comparison to human expert performance. This document, however, focuses on the physical and mechanical properties of the interference screws and their intended use.
Here's a breakdown of why the requested information cannot be extracted from the provided text, followed by what can be extracted:
Why the Requested Information (for an AI/ML Model) Cannot Be Provided:
- No AI/ML Model: The document describes a physical medical device (implants), not a software algorithm or AI/ML model. Therefore, concepts like "acceptance criteria for an AI model," "sample size for test set," "data provenance," "number of experts for ground truth," "adjudication method," "MRMC study," "standalone performance," "training set," or "ground truth for training set" are not applicable to the content provided.
- Focus on Substantial Equivalence: The primary objective of this 510(k) submission is to demonstrate "substantial equivalence" to existing, legally marketed predicate devices. This involves comparing technological characteristics, materials, and indications for use, often supported by non-clinical (bench) testing, rather than clinical efficacy studies in the way an AI/ML device would be evaluated.
Information Extracted from the Provided Text (Relevant to the Medical Device itself):
Since the prompt asks for a table of acceptance criteria and reported device performance, and the document is about a physical device's non-clinical testing, I will interpret "acceptance criteria" as the performance metrics evaluated during the non-clinical testing and "reported device performance" as the implication that these tests demonstrated substantial equivalence. However, specific numerical targets (e.g., "pullout strength > X N") are not provided.
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria (Measured Performance Metric) | Reported Device Performance (as implied by FDA Substantial Equivalence) |
|---|---|
| Insertion Torque | Demonstrated substantial equivalence to predicate devices. |
| Anchor Pullout (at T=0, 3, 6 and 12 week in-vitro physiological aging) | Demonstrated substantial equivalence to predicate devices. |
| Torque to Failure | Demonstrated substantial equivalence to predicate devices. |
Note: The document states, "Results of performance and safety testing have demonstrated that the proposed device is substantially equivalent to the predicate devices." It does not provide specific numerical values for the acceptance criteria or the device's performance, but rather confirms that the device met the necessary standards for substantial equivalence based on these non-clinical tests.
The remaining numbered points (2 through 9) are explicitly related to evaluating an AI/ML model, which this document does not describe. Therefore, the answer for those points is that the information is Not Applicable or Not Provided in the context of this medical device submission.
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
- Not Applicable / Not Provided. This document describes non-clinical (bench) testing of a physical implant, not an AI/ML model evaluated with a test set of data. The "test set" would be the screws themselves and the materials used in the bench tests (e.g., bone surrogates). The sample sizes for these bench tests are not specified, nor is the provenance of the materials used.
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)
- Not Applicable / Not Provided. This concept relates to human expert annotation for AI/ML ground truth. For this physical device, "ground truth" would be the objective results of the mechanical tests, not expert consensus on images or other data.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
- Not Applicable / Not Provided. Adjudication methods are used to resolve disagreements among human experts providing ground truth for AI/ML models. This is not relevant to the non-clinical testing of an orthopedic implant.
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
- Not Applicable / Not Provided. An MRMC study evaluates the performance of human readers, sometimes with and without AI assistance. This document pertains to a physical medical device and does not involve human readers or AI assistance in its evaluation.
6. If a standalone (i.e. algorithm only without human-in-the loop performance) was done
- Not Applicable / Not Provided. This refers to the standalone performance of an AI algorithm. The device is a physical implant, not an algorithm.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
- Objective Mechanical Test Results. For the non-clinical testing described (Insertion Torque, Anchor Pullout, Torque to Failure), the "ground truth" would be the objective, quantifiable measurements obtained from the mechanical tests performed on the implant, not expert consensus or pathology in a clinical context.
8. The sample size for the training set
- Not Applicable / Not Provided. There is no "training set" as this is not an AI/ML model.
9. How the ground truth for the training set was established
- Not Applicable / Not Provided. There is no "training set" or corresponding ground truth establishment process for this physical device.
§ 888.3030 Single/multiple component metallic bone fixation appliances and accessories.
(a)
Identification. Single/multiple component metallic bone fixation appliances and accessories are devices intended to be implanted consisting of one or more metallic components and their metallic fasteners. The devices contain a plate, a nail/plate combination, or a blade/plate combination that are made of alloys, such as cobalt-chromium-molybdenum, stainless steel, and titanium, that are intended to be held in position with fasteners, such as screws and nails, or bolts, nuts, and washers. These devices are used for fixation of fractures of the proximal or distal end of long bones, such as intracapsular, intertrochanteric, intercervical, supracondylar, or condylar fractures of the femur; for fusion of a joint; or for surgical procedures that involve cutting a bone. The devices may be implanted or attached through the skin so that a pulling force (traction) may be applied to the skeletal system.(b)
Classification. Class II.