(90 days)
NEXUS® Hip stems are intended for use in total hip arthroplasty for reduction or relief of pain and/or improved hip function in skeletally mature patients.
- non-inflammatory degenerative joint disease such as osteoarthritis, avascular necrosis, ankylosis, protrusio acetabuli, and painful hip dysplasia;
- inflammatory degenerative joint disease such as rheumatoid arthritis;
- correction of functional deformity; and,
- revision procedures where other treatments or devices have failed
Titanium plasma spray and HA coatings applied to implant surfaces are intended for uncemented arthroplasty.
The MicroPort NEXUS® Hip Stem is manufactured from titanium alloy (Ti6Al4V per ASTM F620) and includes titanium plasma sprayed coating conforming to ASTM F1580 and a hydroxyapatite (HA) coating conforming to ASTM F1185. The proximal body of the hip stem has a dual titanium and HA plasma spray coating while the distal portion of the hip stem has HA coating only. The NEXUS Hip Stem features collared and collarless options and sizes 0 through 12 with standard and high offsets. The NEXUS Hip Stem is provided sterile and is a single use device. The subject NEXUS Stems are compatible with previously cleared MicroPort femoral head and shell/liner components.
This document is a 510(k) clearance letter for the NEXUS® Hip Stem, a medical device. It does not contain information about an AI/ML medical device, clinical study performance data, or details about the establishment of ground truth by experts. The relevant sections for describing device performance are focused on non-clinical (mechanical) testing of the hip stem itself.
Therefore, I cannot fulfill your request for information related to acceptance criteria, study details, expert involvement, and ground truth establishment in the context of an AI/ML device. The provided FDA clearance letter is for a physical medical implant, not an AI software.
The sections you've asked about (acceptance criteria, sample size, expert ground truth, adjudication, MRMC studies, standalone performance, training set details) are typically found in clinical performance studies for AI/ML-driven medical devices, especially those involving image analysis or diagnostic support. This document is instead a regulatory clearance for a traditional orthopedic implant, where "performance" relates to mechanical and material characteristics and "testing" involves physical simulations or analyses, not analysis of clinical data by AI.
To illustrate what you would expect to find if this were an AI/ML device, and why it's not present here:
An AI/ML device would typically have acceptance criteria related to its diagnostic or predictive performance metrics (e.g., sensitivity, specificity, AUC, FROC curves) on a test set. The study proving these criteria would describe:
- A table of acceptance criteria and reported device performance: This would list target thresholds (e.g., "Sensitivity >= 90%") and the actual observed performance (e.g., "Sensitivity = 92.5%").
- Sample size and data provenance: Details on how many cases (e.g., images, patient records) were in the test set, where they came from (e.g., multi-center, specific countries), and if they were collected retrospectively or prospectively.
- Experts for ground truth: The number and qualifications of clinicians (e.g., board-certified radiologists with X years of experience, specializing in Y disease) who reviewed the test cases to establish the definitive diagnosis or finding.
- Adjudication method for ground truth: How disagreements among experts were resolved (e.g., 2-out-of-3 majority vote, third independent expert review, consensus meeting).
- MRMC comparative effectiveness study: If the AI was intended to assist human readers, there would be a study comparing human reader performance with AI assistance vs. without AI assistance. The effect size (e.g., improvement in AUC or reading time) would be reported.
- Standalone performance: The algorithm's performance by itself without human intervention.
- Type of ground truth: Whether it was expert consensus, biopsy-proven pathology, long-term patient outcomes, or a combination.
- Training set size and ground truth establishment: Similar details for the data used to train the AI model.
Since none of this information is available in the provided text, I cannot create the table or answer the specific questions posed using the given input. The input is purely for a physical implant, not an AI product.
§ 888.3353 Hip joint metal/ceramic/polymer semi-constrained cemented or nonporous uncemented prosthesis.
(a)
Identification. A hip joint metal/ceramic/polymer semi-constrained cemented or nonporous uncemented prosthesis is a device intended to be implanted to replace a hip joint. This device limits translation and rotation in one or more planes via the geometry of its articulating surfaces. It has no linkage across-the-joint. The two-part femoral component consists of a femoral stem made of alloys to be fixed in the intramedullary canal of the femur by impaction with or without use of bone cement. The proximal end of the femoral stem is tapered with a surface that ensures positive locking with the spherical ceramic (aluminium oxide, A12 03 ) head of the femoral component. The acetabular component is made of ultra-high molecular weight polyethylene or ultra-high molecular weight polyethylene reinforced with nonporous metal alloys, and used with or without bone cement.(b)
Classification. Class II.