Search Results

Indicated for displaying images of the tracheobronchial tree to aid the physician in guiding endoscopic tools or catheters in the pulmonary tract and to enable marker placement within soft lung tissue. It does not make a diagnosis and is not an endoscopic tool. Not for pediatric use.

The superDimension™ navigation system version 7.2 (V7.2) is a device that guides endoscopic tools to a target in or adjacent to the bronchial tree on a path identified by a previous CT scan. The superDimension™ navigation system V7.2 allows visualization of the target and the interior of the bronchial tree; placement of catheters in the bronchial tree; and placement of radiosurgical and dye markers into soft lung tissue to guide radiosurgery and thoracic surgery.

Covidien Ilc is introducing the superDimension™ navigation system software release V7.2, which is a software modification to the predicate device superDimension navigation system cleared under 510(k) K151376. The V7.2 software includes an optional local registration feature intended to compensate for CT-to-body divergence though incorporation of additional fluoroscopic imaging data taken during the electromagnetic navigation procedure. Local registration is an optional feature and can be used at the physician's discretion.

This document describes a 510(k) premarket notification for the superDimension™ Navigation System V7.2. The device is a software modification to a previously cleared predicate device (K151376). The core of the information provided focuses on the substantial equivalence argument, particularly how the safety and effectiveness are maintained despite the modifications.

Based on the provided text, here's an analysis of the acceptance criteria and the study that proves the device meets them:

Key Takeaway: This submission (K173244) is for a software modification (V7.2) to an existing device. Crucially, no clinical studies were required for this submission. The "studies" proving acceptance are primarily design verification and validation tests, and a demonstration of substantial equivalence to the predicate device.

1. Table of Acceptance Criteria and Reported Device Performance

Given that this is a 510(k) for a software update to an already cleared device, the "acceptance criteria" are predominantly about ensuring the updated device continues to meet the safety and effectiveness standards of the predicate, and that the new feature (local registration) functions as intended without introducing new risks or compromising existing functions.

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

• Sample Size: The document does not specify a numerical "sample size" in terms of patient data or clinical cases for testing. The testing described is primarily engineering and usability validation.
  • For software verification, this would typically involve a comprehensive set of test cases designed to exercise all functionalities, including the new local registration feature. The document indicates "target visualization and marking accuracy, local registration accuracy, software application testing and regression testing." These are likely performed on simulated data, phantom studies, and possibly existing de-identified CT scans.
  • For usability validation, the text mentions "representative users from targeted user groups." No specific number of users is provided.
  • For fiducial marker board testing, no specific sample size is mentioned.
• Data Provenance: Not applicable in the context of clinical studies for this submission. The tests are primarily in-house design verification and validation. If any imaging data were used for software testing, its origin (e.g., country) is not specified, but it's implied to be de-identified or synthetic data suitable for testing.
• Retrospective or Prospective: Not applicable as this was not a clinical study involving patients. The design verification and validation were prospective in nature (performed specifically for this submission).

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

• Number of Experts: For usability validation, the document mentions "qualified bronchoscopists and clinicians." A specific number is not provided.
• Qualifications of Experts: They are described as "qualified bronchoscopists and clinicians." This implies medical professionals with relevant experience in the use of such navigation systems in pulmonary procedures. No specific years of experience are listed.
• Ground Truth Establishment: For the technical software verification tests (accuracy, regression), the "ground truth" would be established by engineering specifications and expected outputs generated by the development team. For usability validation, "ground truth" is established by user feedback and observation against defined usability metrics.

4. Adjudication Method for the Test Set

• Adjudication Method: Not applicable. This was not a multi-reader assessment of subjective image interpretations. For usability testing, observed behaviors and subjective feedback from users would be collected and analyzed, likely against predefined criteria, but not through a formal adjudication process akin to clinical diagnostic studies.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

• MRMC Study: No, an MRMC comparative effectiveness study was not done. The document explicitly states: "Clinical tests were not required to validate the changes to the superDimension™ navigation system V7.2."
• Effect Size: Not applicable.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

• Standalone Performance: The software verification testing (e.g., target visualization accuracy, target marking accuracy, local registration accuracy) could be considered a form of "standalone" evaluation of the algorithm's performance against predefined metrics, though the device is inherently intended for human-in-the-loop use. It assesses the software's computational correctness and accuracy independent of real-time human interaction during a procedure for these specific metrics. However, it's not a standalone diagnostic AI system.

7. The Type of Ground Truth Used

• Type of Ground Truth:
  • For software verification (accuracy): Engineering specifications, mathematically derived correct outputs, or measurements from controlled phantom studies.
  • For usability validation: User feedback, observed user interactions, and conformance to predefined user needs and task completion criteria.
  • Clinical ground truth (e.g., pathology, outcomes data) was not used, as no clinical studies were performed.

8. The Sample Size for the Training Set

• Sample Size for Training Set: Not applicable. This submission is for a software update for a medical device that guides instruments, not for a machine learning/AI algorithm that requires a "training set" in the conventional sense (e.g., for image classification or prediction). The "software modification" refers to changes in the core application logic and potentially new computational features (like local registration), which are developed through traditional software engineering processes, not machine learning model training.

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

• Ground Truth for Training Set Establishment: Not applicable, as there was no machine learning training set in the context of this device update. The "ground truth" for the software's functionality and accuracy would be established through its design specifications and algorithms, which are based on established physics, geometry, and engineering principles relevant to navigation and imaging.

Ask a specific question about this device

Indicated for displaying images of the tracheobronchial tree to aid the physician in guiding endoscopic tools or catheters in the pulmonary tract and to enable marker placement within soft lung tissue. It does not make a diagnosis and is not an endoscopic tool. Not for pediatric use.

The superDimension™ navigation system is a device that guides a bronchoscope and endoscopic tools to a target in or adjacent to the bronchial tree on a path identified by CT scan. The superDimension™ navigation system also allows visualization of the target and the interior of the bronchial tree; placement of catheters in the bronchial tree; and placement of radiosurgical and dye markers into soft lung tissue to guide radiosurgery and thoracic surgery.

This document is a 510(k) premarket notification for the superDimension Navigation System, specifically for software release version 7.1. It details the substantial equivalence to a predicate device and the testing performed.

Here's an analysis of the provided information regarding acceptance criteria and the study that proves the device meets them:

1. A table of acceptance criteria and the reported device performance

The document does not present a formal table of acceptance criteria with corresponding performance metrics in the format typically seen for algorithms with quantitative outputs (e.g., sensitivity, specificity, accuracy against a numerical threshold). Instead, the "acceptance criteria" are implied through the design verification and validation processes.

The reported performance is qualitative and focused on the successful execution of tasks and compliance with specifications:

2. Sample size used for the test set and the data provenance

• Test set sample size: The document does not specify a numerical sample size for the "test set" in terms of patient data or CT scans. The testing described focuses on the software's functionality and user interaction rather than its performance on a dataset of clinical cases. The "test set" for design validation involved representative users (bronchoscopists, clinicians, and service technicians) performing "typical use scenario defined in the design validation protocol." However, the exact number of users is not stated, nor is the number of scenarios or cases they performed.
• Data provenance: Not applicable in the traditional sense of a clinical dataset for an AI algorithm (e.g., retrospective/prospective patient data from specific countries). The validation was conducted under simulated use conditions at Covidien LLC facilities in Plymouth, MN (USA) and Herzliya, Israel.

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

This information is not directly applicable in the context of this device's validation as described. The validation was a functional and user-experience validation, not an assessment of an AI algorithm's diagnostic accuracy against a "ground truth" derived from expert consensus. The "experts" involved were the "representative users" who participated in the design validation:

• Qualifications: "qualified bronchoscopists, clinician, and service technicians."
• Number of experts: The specific number is not provided, only stating "representative users from each targeted user group."

4. Adjudication method for the test set

Not applicable. The validation involved users successfully completing tasks, rather than an adjudication process for a diagnostic outcome. The success was evidently unanimous ("100 % of users... successfully completed validation tasks"), suggesting no need for adjudication for disagreement.

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

No, an MRMC comparative effectiveness study was not conducted or described. The device is a navigation system that aids in guiding tools, not an AI intended to improve human reader diagnostic accuracy (e.g., in reading images). The changes described are software enhancements for planning and procedure, not a new AI diagnostic component.

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

This concept is not directly applicable to the device as described. The superDimension Navigation System is an interactive system designed to be used by a human (physician) to guide tools. It is not a standalone algorithm that provides a diagnosis or output without human interaction. The validation focused on the system's functionality and its successful use by humans.

7. The type of ground truth used

The "ground truth" was the pre-defined successful completion of "typical use scenarios" and conformance to product specifications and user needs. This is essentially a functional and user-acceptance "ground truth" for a medical device software modification, rather than a clinical ground truth like pathology, expert consensus on imaging, or outcomes data, which would be typical for a diagnostic AI.

8. The sample size for the training set

The document does not describe a "training set" in the context of machine learning or AI models. The software modifications are described as "software features" and "minor software enhancements," suggesting traditional software development and testing, not a machine learning model that requires a training set.

9. How the ground truth for the training set was established

Not applicable, as no training set for a machine learning model is mentioned.

Ask a specific question about this device