K Number
K234056
Date Cleared
2024-04-24

(124 days)

Product Code
Regulation Number
888.3600
Panel
OR
Reference & Predicate Devices
N/A
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

The canturio® se (Canturio Smart Extension) with Canary Health Implanted Reporting Processor (CHIRP) System is intended to provide objective kinematic data from the implanted medical device during a patient's total knee arthroplasty (TKA) post-surgical care. The kinematic data are an adjunct to other physiological parameter measurement tools applied or utilized by the physician during patient monitoring and treatment post-surgery.

The device is indicated for use in patients undergoing a cemented TKA procedure that are normally indicated for at least a 30mm sized tibial stem extension.

The objective kinematic data generated by the canturio® se with CHIRP System are not intended to support clinical decision-making and have not been shown to provide any clinical benefit.

The canturio® se with CHIRP System is compatible with Zimmer Personalized Knee System.

Device Description

The canturio® se is a tibial extension implant or stem that is attached by the orthopedic surgeon to the Zimmer Biomet Persona® tibial baseplate to form the patient's tibial knee prosthesis. The software and electronics embedded within the canturio® se prosthesis collect the patient's functional movement and gait parameter data post-surgery. Like a traditional tibial extension, the canturio® se provides additional stability in the same manner as a traditional knee extension.

The canturio® se (Canturio Smart Extension or CSE) is used with the Canary Health Implanted Reporting Processor (CHIRP) System which is comprised of the following subsystems:

  • Operating Room Base Station Subsystem ("OR BS"),
  • Home Base Station Subsystem (“HBS”),
  • Canary Cloud Data Management Platform ("Cloud" or "CMDP") subsystem and
  • Canary Medical Gait Parameters (CMGP) software module.
    Each CHIRP subsystem combines physical components, electronics, software, and user interfaces to collect, store, analyze, transmit, and display patient data for use by both physicians and patients.
AI/ML Overview

The provided text describes a 510(k) premarket notification for a medical device called "canturio® se (Canturio Smart Extension)". This document focuses on demonstrating substantial equivalence to a predicate device, rather than providing a detailed report of a study proving the device meets specific acceptance criteria for AI performance.

Therefore, many of the requested details, such as the specific acceptance criteria for AI/algorithm performance, ground truth establishment methods, expert qualifications, and MRMC study results, are not explicitly available in the provided text, as this is a traditional medical device submission, not specifically one for an AI/ML medical device where those details would be paramount.

However, based on the information available, here's what can be extracted and inferred:

Device Purpose and General Performance Testing:

The canturio® se with Canary Health Implanted Reporting Processor (CHIRP) System is intended to provide objective kinematic data from the implanted medical device during a patient's total knee arthroplasty (TKA) post-surgical care. The kinematic data are an adjunct to other physiological parameter measurement tools. It's crucial to note the statement: "The objective kinematic data generated by the canturio® se with CHIRP System are not intended to support clinical decision-making and have not been shown to provide any clinical benefit." This indicates that the primary function is data collection and not an AI-driven diagnostic or treatment recommendation system that would typically require extensive performance criteria and studies as outlined in your request.

The performance data summarized focuses on non-clinical tests to support modifications to the device itself, ensuring its safety and functionality as an implantable device that collects data. These tests are not "AI performance" related in the sense of accuracy, sensitivity, specificity, etc., for a machine learning model.

Acceptance Criteria and Reported Device Performance (Inferred from device type and summary):

Since the device is stated to provide "objective kinematic data" and is a "product line extension" of a predicate device, the acceptance criteria would likely revolve around the accuracy, reliability, and reproducibility of these kinematic measurements themselves. The "reported device performance" would be the successful demonstration of these factors through non-clinical testing.

Acceptance Criteria (Inferred)Reported Device Performance (From "Summary of Performance Data")
Mechanical Integrity/Durability: Device withstands expected mechanical stresses and fatigue over its intended lifespan.Mechanical Fatigue: Tested to ensure durability. (Passed/Met - inferred by clearance)
Electrical Integrity/Longevity: Electronic components function reliably for the device's lifespan; battery performance meets specifications.Shock Survival: Tested. (Passed/Met - inferred by clearance)
Electrical Life Test: Performed. (Passed/Met - inferred by clearance)
Battery Longevity: Tested. (Passed/Met - inferred by clearance)
Electronic Functionality: Tested. (Passed/Met - inferred by clearance)
Accuracy, Reliability, and Reproducibility of Kinematic Measurements: The data collected by the device is consistent and accurate.Accuracy, reliability, and reproducibility of kinematic measurements: Tested. (Passed/Met - inferred by clearance)
Electromagnetic Compatibility (EMC) and Interference (EMI): Device operates without problematic interference and is not susceptible to external EMI.Electromagnetic compatibility (EMC) and electromagnetic interference (EMI): Tested. (Passed/Met - inferred by clearance)
Biocompatibility: Materials are safe for implantation.Biological Evaluation: Performed. (Passed/Met - inferred by clearance)
Hermeticity: Electronic enclosures protect components from bodily fluids.Hermeticity of any electronic component enclosures: Tested. (Passed/Met - inferred by clearance)
Magnetic Resonance Compatibility: Device is safe for use with MRI.Magnetic Resonance Compatibility: Tested. (Passed/Met - inferred by clearance)
Software Functionality: Software components (OR BS, HBS, Cloud, CMGP) perform as intended.Software Verification: Performed. (Passed/Met - inferred by clearance)

Since this is a 510(k) for a hardware device that collects data, and explicitly states the data is not for clinical decision-making or benefit, the traditional AI/ML performance study criteria listed in your request are not applicable or detailed in this document. The focus for this review is substantial equivalence to the predicate device based on material, design, and general device performance.

Therefore, for the remainder of your points, the answer is:

  1. Sample sizes used for the test set and the data provenance: Not specified for "kinematic measurements" in terms of subject count, nor is data provenance (country of origin, retrospective/prospective) for such measurement described. The testing described appears to be primarily bench/laboratory non-clinical testing of the device hardware and software.
  2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable/not specified. The "ground truth" for this device would be objective physical measurements, not expert interpretations.
  3. Adjudication method (e.g. 2+1, 3+1, none) for the test set: Not applicable/not specified.
  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. This device is not an AI interpretation tool for human readers; it's a data collection device.
  5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: The "kinematic measurements" functionality is "standalone" in that it captures data without human input beyond the system setup. However, this isn't an AI algorithm in the sense of a diagnostic or predictive model.
  6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.): For kinematic measurements, the ground truth would typically be established by highly accurate motion capture systems or physical measurement standards, not expert consensus or pathology.
  7. The sample size for the training set: Not applicable. This device is not described as having an AI/ML model that requires a "training set" in the conventional sense for a diagnostic or prognostic application. The "software verification" refers to traditional software validation processes.
  8. How the ground truth for the training set was established: Not applicable, as there's no described AI/ML training set.

§ 888.3600 Implantable post-surgical kinematic measurement knee device.

(a)
Identification. An implantable post-surgical kinematic measurement knee device is a device that provides objective kinematic data after total knee arthroplasty surgery. The kinematic data provided by the device are used as an adjunct to other physiological parameter measurement tools utilized during the course of patient monitoring and treatment post surgery.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Non-clinical performance testing must demonstrate that the device performs as intended under anticipated conditions of use. The following tests must be conducted:
(i) Mechanical testing must evaluate the mechanical function (mechanical fatigue, static mechanical strength) and durability of the implant.
(ii) Simulated use testing must evaluate the ability of the device to be sized, inserted, and sufficiently secured to any compatible components.
(iii) Testing must demonstrate the accuracy, reliability, and reproducibility of kinematic measurements.
(iv) Testing must demonstrate diagnostic and therapeutic ultrasound conditions for safe use.
(v) Testing must demonstrate that the device performs as intended under anticipated conditions of use demonstrating the following performance characteristics, if applicable:
(A) Magnetic pulse output testing;
(B) Magnetic and electrical field testing; and
(C) Testing of the safety features built into the device.
(vi) Testing must demonstrate hermeticity of any electronic component enclosures.
(2) Performance testing must evaluate the compatibility of the device in a magnetic resonance (MR) environment.
(3) Human factors testing must demonstrate that the intended user(s) can correctly use the device for its intended use, including for implantation and post-procedure data access.
(4) Performance data must demonstrate the sterility of the device implant and patient-contacting components.
(5) Performance data must validate the reprocessing instructions for the reusable components of the device.
(6) The patient-contacting components of the device must be demonstrated to be biocompatible.
(7) Design characteristics of the device, including engineering schematics, must ensure that the geometry and material composition are consistent with the intended use.
(8) Performance testing must demonstrate the electromagnetic compatibility/interference, (EMC/EMI), electrical safety, thermal safety, battery safety, and wireless performance of the device.
(9) Software verification, validation, and hazard analysis must be performed.
(10) The labeling must include the following:
(i) A shelf life;
(ii) Physician and patient instructions for use, including images that demonstrate how to interact with the device;
(iii) Detailed instruction of the surgical technique;
(iv) Hardware and software requirements for interacting with the device;
(v) A clear description of the technological features of the device including identification of the device materials, compatible components, and the principles of operation;
(vi) Identification of magnetic resonance (MR) compatibility status;
(vii) Validated methods and instructions for reprocessing of any reusable components; and
(viii) A statement regarding the limitations of the clinical significance of the kinematic data.