Search Results

The Restoris Partial Knee Application (PKA), for use with the Robotic Arm Interactive Orthopedic System (RIO), is intended to assist the surgeon in providing software defined spatial boundaries for orientation and reference information to anatomical structures during orthopedic procedures.

The Restoris Partial Knee Application (PKA), for use with the Robotic Arm Interactive Orthopedic System (RIO), is indicated for use in surgical knee procedures, in which the use of stereotactic surgery may be appropriate, and where reference to rigid anatomical bony structures can be identified relative to a CT based model of the anatomy. These procedures include unicondylar knee replacement and/or patellofemoral knee replacement.

Restoris PKA is an upgrade to the Tactile Guidance System v2.0, a.k.a RIO, which was cleared via K081867. The features of this application are to improve overall performance of the system in supporting unicondylar and/or patellofemoral knee replacement. Restoris PKA is used with RIO which includes an optical detector, robotic arm, and guidance module. In addition, the application is designed to be used with a pre-operative planning laptop, as well as both reusable and disposable instrumentation.

Restoris PKA provides stereotactic guidance during minimally invasive orthopedic surgical procedures by using patient CT data to assist a surgeon with presurgical planning and interpretive/intraoperative navigation. RIO's robotic arm, once configured for use with Restoris PKA, can serve as a surgeon's "intelligent" tool holder or tool guide by passively constraining the preparation of an anatomical site for an orthopedic implant with softwaredefined spatial boundaries.

Here's a breakdown of the acceptance criteria and study information for the Restoris Partial Knee Application (PKA) as described in the provided 510(k) summary:

1. Table of Acceptance Criteria and Reported Device Performance

The provided document summarizes the testing and indicates that "All acceptance criteria were satisfied" or "The testing successfully met the acceptance criteria specified in the validation protocol." However, the specific quantitative acceptance criteria themselves are not explicitly listed in this summary document. The document focuses on the conclusion that the criteria were met.

2. Sample Sizes Used for the Test Set and Data Provenance

• Verification Testing (Integration & Registration):

  • Sample Size: Sawbone models were used for both the integration and registration verification tests. The exact number of sawbone models is not specified.
  • Data Provenance: Retrospective (synthetic models).

• Validation Testing (Usability):

  • Sample Size: Two (2) cadaveric specimens were used.
  • Data Provenance: Prospective (cadaveric specimens).

• Validation Testing (Overall System Performance):

  • Sample Size: Three (3) cadaveric specimens were used.
  • Data Provenance: Prospective (cadaveric specimens).

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

• Verification Testing: Not applicable, as testing was against predefined system specifications or expected outcomes.
• Validation Testing (Usability): Three (3) users participated in completing the procedure. Their specific qualifications (e.g., "radiologist with 10 years of experience") are not explicitly stated. They are generally referred to as "users," implying they are likely surgeons or trained medical professionals using the device in a simulated environment.
• Validation Testing (Overall System Performance): One surgeon performed the procedure, and four (4) independent reviewers assessed the outcome. The specific qualifications of these surgeons/reviewers are not explicitly stated.

4. Adjudication Method for the Test Set

• Verification Testing: Not applicable, as outcomes were likely assessed against objective technical specifications.
• Validation Testing (Usability): The method of adjudication for the "acceptance criteria" being met is not detailed beyond the conclusion that they were met. It's implied to be based on the experience of the 3 users.
• Validation Testing (Overall System Performance): One surgeon performed the procedure, and four (4) independent reviewers assessed it. The specific adjudication method among the four reviewers (e.g., 2+1, 3+1, simple majority) is not specified. It just states they independently reviewed the procedure.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done, What was the Effect Size of How Much Human Readers Improve with AI vs Without AI Assistance

• A MRMC comparative effectiveness study, comparing human readers with and without AI assistance, was not explicitly described in this 510(k) summary. The studies described focus on the device's performance in guiding a surgeon directly during a procedure, not on a diagnostic reading task. The device assists the surgeon rather than providing interpretations for readers.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was Done

• Yes, in part. The Verification Testing for "registration with the Restoris PKA satisfies the specified accuracy requirements" using "the software's improved bone model imaging algorithm" can be considered a form of standalone testing for a specific algorithmic component (registration accuracy). However, the overall device (Restoris PKA) is intended for human-in-the-loop surgical assistance, so the primary validation is in that context.

7. The Type of Ground Truth Used

• Verification Testing (Integration & Registration): The ground truth was likely based on pre-defined engineering or system specifications and benchmarks for integration functionality and accuracy of registration.
• Validation Testing (Usability & Overall System Performance): The ground truth for these cadaveric studies would be the successful and accurate completion of the simulated surgical procedure as assessed by the participating surgeons/reviewers against clinical standards and the stated "needs of the user" for unicondylar and/or patellofemoral knee replacement. This is a form of expert assessment of procedural correctness and outcome.

8. The Sample Size for the Training Set

• The 510(k) summary does not provide any information regarding a specific training set or its sample size for the Restoris PKA. This document describes verification and validation testing, not the development or training of an AI model in the conventional sense. The device "assists a surgeon with presurgical planning and interpretive/intraoperative navigation" using "patient CT data," implying it's a guidance system built upon established biomechanical models and image processing, rather than a deep learning model requiring a distinct training dataset for its core function.

9. How the Ground Truth for the Training Set Was Established

• As no training set is described, there's no information on how its ground truth might have been established.

Ask a specific question about this device

The Robotic Arm Interactive Orthopedic System (RIO) is intended to assist the surgeon in providing software defined spatial boundaries for orientation and reference information to anatomical structures during orthopedic procedures.

The RIO is indicated for use in surgical knee and hip procedures, in which the use of stereotactic surgery may be appropriate, and where reference to rigid anatomical bony structures can be identified relative to a CT based model of the anatomy. These procedures include:

• Unicondylar knee replacement and/or patellofemoral knee replacement
• Total hip arthroplasty (THA)

The main RIO platform includes an optical detector, computer, dedicated instrumentation, operating software, tools and accessories, drill system, and a robotic arm. The system's architecture is designed to support two main surgical applications: knee procedures (per the predicate device K081867) and THA procedures (per RIO-THA described in this 510(k) submission). With application specific hardware and software, it provides stereotactic guidance during minimally invasive orthopedic surgical procedures by using patient CT data to assist a surgeon with presurgical planning and interpretive/intraoperative navigation. RIO's robotic arm, once configured for a specific application (knee or hip), can serve as surgeon's "intelligent" tool holder or tool guide by passively constraining the preparation of an anatomical site for an orthopedic implant with software-defined spatial boundaries.

The provided document describes the MAKO Surgical Corp.'s Robotic Arm Interactive Orthopedic System - THA (RIO - THA). While it confirms performance testing was done, it does not provide explicit acceptance criteria with specific numerical targets, nor does it detail a study that directly proves the device meets such criteria with quantitative results.

However, based on the provided text, I can infer the general nature of the "acceptance criteria" and the type of study conducted.

Here's an attempt to structure the information as requested, highlighting where specific details are missing:

1. Table of Acceptance Criteria and Reported Device Performance

Note: The document states that the "results of these tests satisfied all required acceptance criteria." However, it does not provide the specific quantitative acceptance criteria themselves (e.g., "accuracy within X mm") or the numerical results achieved by the device.

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Size: Not explicitly stated. The document mentions "sawbone models" and "cadaveric material" without specifying the number of models or cadavers used.
• Data Provenance: The tests were performed "in the laboratory" using "sawbone models" and "cadaveric material." This indicates a pre-clinical, prospective study design conducted in a simulated or ex-vivo environment. The country of origin for the data is not specified but is likely the USA given the submitter's address.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications

• Number of Experts: Not specified.
• Qualifications of Experts: Not specified. The document only mentions that "post-operative x-rays were obtained and evaluated." It does not clarify who performed this evaluation or their credentials.

4. Adjudication Method for the Test Set

• Adjudication Method: Not specified. It's unclear how agreement or disagreement on the "evaluation" of the post-operative x-rays was resolved or if multiple evaluators were involved.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• Was an MRMC study done? No, an MRMC comparative effectiveness study is not mentioned. The study focused on the system's performance (accuracy, functionality, intended use validation) rather than comparing human reader performance with and without AI assistance.
• Effect Size of Human Readers' Improvement with AI vs. Without AI Assistance: Not applicable, as no MRMC study was conducted.

6. Standalone (i.e., algorithm only without human-in-the-loop performance) Study

• Was a standalone study done? Yes, the performance data described appears to be a standalone evaluation of the RIO-THA system's capabilities in a laboratory setting. The "system level verification testing" on sawbones and cadavers directly assesses the device's accuracy and functionality in supporting THA. The "robotic arm... can serve as surgeon's 'intelligent' tool holder or tool guide by passively constraining the preparation of an anatomical site," implying its standalone functionality in guiding surgical actions.

7. Type of Ground Truth Used

• Ground Truth Type:
  • For the sawbone model testing: Likely defined by measurements against engineering specifications or pre-determined anatomical targets through the system's own output or independent measurement tools.
  • For the cadaveric material testing: "Post-operative x-rays were obtained and evaluated in order to validate the system's intended use." This suggests the ground truth was based on radiological assessment after the procedure to confirm the accuracy of implant placement or anatomical preparation according to surgical plans. This is a form of expert consensus (radiological evaluation) of the post-operative outcome.

8. Sample Size for the Training Set

• Training Set Sample Size: Not mentioned. The document describes verification testing but does not detail any machine learning component requiring a distinct training set. The RIO system uses "patient CT data to assist a surgeon with presurgical planning and interpretive/intraoperative navigation," suggesting it processes CT data, but it doesn't indicate a machine learning model that would require a "training set" in the conventional sense of AI development. It might use statistical models or pre-programmed algorithms based on anatomical datasets, but these are not explicitly described as having a "training set."

9. How the Ground Truth for the Training Set Was Established

• How Ground Truth for Training Set Was Established: Not applicable, as no explicit training set for a machine learning model is mentioned or implied. If the system uses pre-programmed anatomical models or statistical data, the ground truth for those would typically be established through extensive anatomical studies, medical imaging databases, and expert anatomical landmarking, but these specifics are not provided.

Ask a specific question about this device