(20 days)
Cube Navigator is a non-invasive software intended for planning the needle trajectory of CT-guided, percutaneous punctures using DICOM-images and a Navigation Cube attached to a patient. It is indicated for the planning the needle trajectory of CT-guided, percutaneous punctures for therapeutic and diagnostic purposes. The software's functionality allows to automatically recognize orientation, position and type of the Navigation Cube present in the DICOM-images and to manually define virtual needle trajectories to a physician-identified target through the Navigation Cube. Based on this planning, Cube Navigator provides physician information for such purpose.
Cube Navigator can only be used in combination with a Navigation Cube, specifically, either the Access Cube for instruments 10-20G (article number mt-ac-ct-20, Class 1 exempt, product code GDF) or the Puncture Cube for instruments 18-22G (article number mt-pc-ct-25, Class 1 exempt, product code GDF). Cube Navigator is a standalone software that integrates the CE-marked and FDA-cleared diagnostic Medical Imaging Viewer medDream manufactured by Softneta that provides all DICOM-related functionality like displaying and manipulating DICOM-images.
The Cube Navigator is intended for interventional radiologists, who perform CT-guided interventions for therapeutic and diagnostic purposes. The intended use environment is the CT control room, which can be in a hospital or a medical office.
Cube Navigator is a non-invasive medical device software used in interventional radiology for planning the needle trajectory of CTguided, percutaneous punctures. Cube Navigator provides planning functionality for needle placement through a separate device, called a Navigation Cube. Cube Navigator allows the user to register the Navigation Cube in a feasible needle traiectory using a virtual needle. Based on this planned trainctor displays needle depth and the enty points on the Navigation Cube in its user interface. The Cube Navigator software is only used for planning the needle trajectory, but requires a CT-scan with a Navigation Cube mounted on a patient for the purposes of visualization.
Navigation Cubes are sterile, disposable needle quides which are attached to the planning scan is acquired. Currently two variants exist: Puncture Cube or Access Cube, both Class 1 exempt, product code GDF. The Navigation Cubes support the use of needle sizes 10-20G (Access Cube) or 18-22G (Puncture Cube), Needle length is dependent on the trained from the displayed needle depth. Cube Navigator detects which Navigation Cube is present in the takes the average of the supported needle diameters for that Navigation Cube when calculating the thickness of the virtual needle.
Cube Navigator is a standalone software that integrates with the CE-marked and FDA-cleared diagnostic Medical Imaging Viewer medDream manufactured by Softneta (Class II, K22232), see Software Architecture Design Cube Navigator for integrated version). Gube Navigator allows the user to access the functionality of medicing images locally or through a remote DICOM-connection, by viewing and manipulating DICOM-images through the user interfaces of the Cube Navigator, as well as accessing the functionality of Cube Navigator itself. The DICOM integration is part of the base software package medDream (see the Statement recarding the DICOM-Standard - Cube Navigator for detailed information about DICOM conformity).
Cube Navigator as well as the Navigation Cubes are manufactured by Medical Templates AG. medDream is manufactured by Softneta and for the described purpose, distributed by Medical Templates AG as bundle together with Cube Navigator. When the user installs Cube Navigator, medDream is also installed, without any additional installation or configuration necessary.
Here's an analysis of the acceptance criteria and study that proves the device meets the acceptance criteria, based on the provided FDA 510(k) summary for Cube Navigator:
1. A table of acceptance criteria and the reported device performance
The provided document does not explicitly list formal "acceptance criteria" with quantitative thresholds for performance. Instead, it describes verification methods and their outcomes. However, we can infer the implicit criteria from the tests performed.
| Acceptance Criteria (Inferred) | Reported Device Performance |
|---|---|
| Automated Registration Accuracy: The ability of the device to automatically detect and register Navigation Cubes in DICOM images. (Implicit criterion: High accuracy) | "Data from phantom studies and retrospectively collected patient cases are used to determine the accuracy of the registration tool to automatically detect and register Navigation Cubes." (While no specific numerical accuracy is given, the statement implies successful determination and meeting of requirements, as the conclusion states the device is "as safe, as effective and performs as well as the predicate device.") |
| Accuracy of Displayed Needle Depth: The device accurately calculates and displays the needle length in the sidebar. (Implicit criterion: Low error in indicated needle length) | "The calculated needle-depth is verified by measuring the tip of the needle to the corresponding plate in the cube with the line measurement tool provided by MedDream (FDA certified product). By comparing the two values to each other the error of the indicated needle length is determined." (Again, no specific numerical error is given, but the overall conclusion suggests acceptable performance.) |
| Accuracy of Virtual Needle Path Planning (Side-by-side comparison): The software's suggested coordinates for a planned virtual needle path coincide with actual physical needle coordinates in retrospective cases. (Implicit criterion: High concordance between planned and actual coordinates) | "For each cube 5 datasets from different tests or studies have been selected and the above conducted process has shown that for all cases the needle coordinates suggested by the software coincide with the actual coordinates for the physical setup." |
| Usability and Safety/Effectiveness: The device is safe and effective for its intended use by interventional radiologists. (Implicit criterion: Positive user feedback and successful use for planning) | "Usability study was conducted where 5 interventional radiologists from independent sites used the software to plan a intervention successfully and provided feedback along the line of the system. It was shown that the Cube Navigator is safe and effective to use." |
2. Sample sizes used for the test set and the data provenance
- Automated Registration Accuracy: "Data from phantom studies and retrospectively collected patient cases." The specific number of phantom studies or patient cases is not disclosed. The country of origin is not disclosed. The data provenance is stated to be retrospective for patient cases, and phantom studies (which are typically prospective/controlled).
- Accuracy of Displayed Needle Depth: "a dataset is loaded." The specific number of datasets is not disclosed. Provenance is not disclosed.
- Accuracy of Virtual Needle Path Planning: "For each cube 5 datasets from different tests or studies have been selected." Since there are two types of cubes (Access Cube and Puncture Cube), this implies a sample size of at least 10 datasets (5 for Access Cube, 5 for Puncture Cube) for this test. These were described as "retrospective" on "existing data." The country of origin is not disclosed.
- Usability Study: 5 interventional radiologists used the software. The provenance of the patient cases or phantoms used during this usability study is not disclosed, but the study itself (user interaction) is prospective.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- For the "Accuracy of Displayed Needle Depth" test, the "line measurement tool provided by MedDream (FDA certified product)" was used, implying a software-based ground truth measurement.
- For the "Accuracy of Virtual Needle Path Planning" test, the comparison was made against "actual coordinates used for the physical needle." This suggests a physical ground truth, likely established during the original acquisition of the "existing data," but not further detailed as involving expert consensus specific to this study.
- The usability study involved 5 interventional radiologists as users, and their feedback contributed to the "ground truth" regarding usability. Their qualifications are stated as "interventional radiologists," but no further details such as years of experience are provided.
4. Adjudication method for the test set
The document does not explicitly describe an adjudication method with multiple readers or a consensus process for establishing ground truth for the core performance tests (registration accuracy, needle depth, path planning). The tests appear to rely on objective measurements (e.g., comparison to existing physical coordinates, measurements with a certified tool, or an implied internal standard for automatic registration).
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
There is no indication that a multi-reader multi-case (MRMC) comparative effectiveness study was performed or that human readers' performance with and without AI assistance was evaluated. The studies described are primarily focused on the software's standalone accuracy and usability, rather than its impact on human reader performance.
6. If a standalone (i.e. algorithm only without human-in-the loop performance) was done
Yes, the described tests for "Automated Registration Accuracy," "Accuracy of Displayed Needle Depth," and "Accuracy of Virtual Needle Path Planning" appear to be standalone (algorithm only) performance evaluations. For example, the automatic detection/registration is an algorithm-only function, and the calculation of needle depth/coordinates also relies on the algorithm. While human interaction is required to use the software (planning the trajectory), the accuracy of the software's output is what's being evaluated in these specific tests, separating it from the human's contribution to the overall clinical outcome.
7. The type of ground truth used
The types of ground truth used include:
- Physical/Retrospective Data: For virtual needle path planning, comparison to "actual coordinates used for the physical needle" from existing data.
- Software-Certified Tool: For displayed needle depth, comparison to measurements made with "MedDream (FDA certified product)."
- Phantom Studies: For automated registration accuracy, implicitly involving known phantom geometries.
- User Feedback: For usability, "feedback along the line of the system" from interventional radiologists.
8. The sample size for the training set
The document does not disclose the sample size for the training set used for the Cube Navigator's algorithms (if any AI/ML components were trained).
9. How the ground truth for the training set was established
Since the training set size is not disclosed, the method for establishing its ground truth is also not disclosed.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).