(351 days)
FastStroke is a CT image analysis software package that assists in the analysis and visualization of CT data derived from DICOM 3.0 compliant CT scans. FastStroke is intended for the purpose of displaying vasculature of the head and neck at different time points of enhancement.
The software will assist the user by providing optimized display settings to enable fast review of the images in synchronized formats, aligning the display of the order of the scans and linking together multiple groups of scans. In addition, the software fuses the vascular information from different time points into a single colorized view. This multiphase information can aid the physician in visualizing the presence or absence of collateral vessels in the brain. Collateral vessel information may aid the physician in the evaluation of stroke patients.
CT perfusion 4D is an image analysis software package that allows the user to produce dynamic image data and to generate information with regard to changes intensity over time. It supports the analysis of CT perfusion images (in the head and body) after the intravenous injection of contrast, in calculation of the various perfusion-related parameters (i.e. regional blood volume, mean transit time and capillary permeability). The results are displayed in a user-friendly graphic format as parametric images.
This software will aid in the assessment of the extent and type of perfusion, blood volume and capillary permeability changes, which may be related to stroke or tumor angiogenesis and the treatment thereof.
NeuroPackage is a solution which contains two medical devices FastStroke and CT Perfusion 4D (Neuro) in order to help streamline the CT Stroke Workflow. The configuration of NeuroPackage enables the user to open a single application, FastStroke, which provides them access to both the updated CT Perfusion 4D and FastStroke applications. However, same as the predicate devices, the capabilities in CT Perfusion 4D and FastStroke can be offered independently.
CT perfusion 4D is an image analysis software package, which allows the user to produce dynamic image data and to generate information with regards to changes in image intensity over time. It supports the analysis of CT Perfusion images (in the head and body) after the intravenous injection of contrast, and calculation of the various perfusion-related parameters (i.e. regional blood flow, regional blood volume, mean transit time and capillary permeability). The results are displayed in a user-friendly graphic format as parametric images.
This software will aid in the assessment of the extent and type of perfusion, blood volume, and capillary permeability changes, which may be related to stroke or tumor angiogenesis and the treatment thereof.
FastStroke is a CT image analysis software package intended for the purpose of displaying stroke workup images (i.e. vasculature of the head, non-contrast head and neck at different time points of enhancement) in a single software, using an optimized workflow. The software is compatible with DICOM 3.0 images and will assist the user by providing dedicated review steps and optimized display settings to enable fast review of the images in synchronized formats. In addition, if a multiphase CT Angiogram has been acquired, the software will fuse the vascular information from these different time points into a single colorized view. This multiphase information can aid the physician in visualizing the presence or absence of collateral vessels in the brain, as well as their delay.
All features from the CT Perfusion 4D Neuro software are accessible in the Perfusion, Set Symmetry and Tissue Classification steps within FastStroke Application, seamlessly introduced in an integrated workflow.
The provided text describes the 510(k) premarket notification for GE Medical Systems SCS's FastStroke and CT Perfusion 4D devices. It highlights the comparison between the proposed devices and their predicate devices.
Here's an analysis of the acceptance criteria and study information, based solely on the provided text. Many aspects of a comprehensive study are not detailed in this summary, as is common in 510(k) executive summaries.
Acceptance Criteria and Reported Device Performance
The acceptance criteria are not explicitly listed in a quantitative table with specific thresholds (e.g., "accuracy > 90%"). Instead, the document states general qualitative criteria and outcomes, primarily focusing on proving substantial equivalence to predicate devices.
The key acceptance criterion described for the brain ventricle segmentation deep learning algorithm is:
- "bench tests that compare the output of the new algorithm with ground truth annotated by qualified experts show that the algorithm performed as expected."
The "performed as expected" is a qualitative statement of acceptance. No specific quantitative performance metric (e.g., Dice score, precision, recall, accuracy) or threshold is provided.
Table of Acceptance Criteria and Reported Device Performance:
| Feature/Metric | Acceptance Criteria (Stated or Implied) | Reported Device Performance (as per text) |
|---|---|---|
| CT Perfusion 4D: Brain Ventricle Segmentation | Output of the new algorithm performs "as expected" when compared to ground truth. | "bench tests... show that the algorithm performed as expected." |
| FastStroke: All other changes (including email feature) | No new questions of safety and effectiveness compared to predicate device. | "The software testing and the corresponding results... did not raise new questions of safety and effectiveness." |
| CT Perfusion 4D: All other changes (incl. email, tissue classification) | No new questions of safety and effectiveness compared to predicate device. | "The software testing and the corresponding results... did not raise new questions of safety and effectiveness." |
| Overall substantial equivalence | Device is as safe and effective as predicate devices. | "considered... to be as safe, as effective as the predicate devices, and is substantially equivalent." |
Study Information:
-
Sample Size Used for the Test Set and Data Provenance:
- Test Set Sample Size: Not specified. The text only mentions "bench tests."
- Data Provenance: Not specified (e.g., country of origin, retrospective/prospective).
-
Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:
- Number of Experts: Not specified. The text mentions "qualified experts" (plural).
- Qualifications of Experts: Not specified beyond "qualified experts." (e.g., specialty, years of experience are not mentioned).
-
Adjudication Method for the Test Set:
- Adjudication Method: Not specified. The general wording "ground truth annotated by qualified experts" does not describe an adjudication process (e.g., consensus, majority rule).
-
Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- Was an MRMC study done? No. The document does not describe any study involving human readers or comparative effectiveness with and without AI assistance for either FastStroke or CT Perfusion 4D. The focus appears to be on the performance of the algorithm itself and a comparison to previous versions of the software.
- Effect Size of Human Reader Improvement: Not applicable, as no MRMC study was described.
-
Standalone Performance (Algorithm Only without Human-in-the-Loop Performance):
- Was a standalone study done? Yes, implicitly. The "bench tests" of the brain ventricle segmentation deep learning algorithm, where its "output" was compared to ground truth, describe a standalone evaluation of the algorithm's performance. The statement "The modified CT Perfusion 4D software employs a deep learning convolutional network to segment the brain ventricles while the predicate device uses a manual method based on HU thresholding" further supports this.
-
Type of Ground Truth Used:
- Brain Ventricle Segmentation: "ground truth annotated by qualified experts." This suggests expert consensus or annotation rather than pathology or outcomes data.
- Other changes: For other changes, it's implied that the "ground truth" for showing non-inferiority was based on comparison to the established performance and safety of the predicate devices, likely through verification and validation tests rather than an independent "ground truth" dataset for clinical outcomes.
-
Sample Size for the Training Set:
- Training Set Sample Size: Not specified. This information is typically proprietary or not included in summary documents.
-
How Ground Truth for the Training Set Was Established:
- Ground Truth Establishment for Training: Not specified. It's common for ground truth for training data to be established similarly to test data (e.g., expert annotation), but this document does not explicitly state it.
§ 892.1750 Computed tomography x-ray system.
(a)
Identification. A computed tomography x-ray system is a diagnostic x-ray system intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data from the same axial plane taken at different angles. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II.