(114 days)
The VIDA|vision software provides reproducible CT values for pulmonary tissue, which is essential for providing quantitative support for diagnosis and follow up examinations. VIDA|vision can be used to support the physician in the diagnosis and documentation of pulmonary tissue images (e.g., abnormalities) from CT thoracic datasets. Three-D segmentation and isolation of sub-compartments, volumetric analysis, density evaluations, low density cluster analysis and reporting tools are combined with a dedicated workflow. The VIDA vision software package is also intended to be a real-time interactive evaluation in space and time for CT volume data sets that provides the reconstruction of two dimensional images into a three-dimensional image format.
VIDA|vision is a self-contained image analysis software package. This real-time interactive evaluation in space and time of CT volume datasets provides the reconstruction of two-dimensional images into a three-dimensional image format.
VIDA|vision can be used to support the physician in the diagnosis, treatment planning, and documentation of chest diseases, including lung cancer, asthma, COPD, interstitial lung disease and other lung abnormalities e.g. when examining the pulmonary and thoracic tissue (i.e. lung parenchyma) in CT thoracic datasets.
Evaluation (3D segmentation & isolation of sub-compartments, volumetric analysis, density evaluations, and low density cluster analysis), editing, and reporting tools are combined with a dedicated workflow.
VIDA|vision is designed to analyze pulmonary CT slice data and display analysis results. Each voxel of the scan is measured by Hounsfield units (HU), a measurement of x-ray attenuation that is applied to each volume element in three dimensional space ("voxel"). The HU are utilized to distinguish between air, water, tissue and bone, such distinction is common in the industry.
VIDA|vision provides computed tomography (CT) viewing, airway analysis, and parenchymal density analysis in one application. VIDA|vision provides imaging of bronchial airways that can be used to assess therapy effectiveness and treatment plan based on CT scan data. VIDA|vision reconstructs multiple cross-section images from CT data into a computer model displaying complex bronchial branches.
VIDA|vision provides quantitative measurements and tabulates quantitative properties. VIDA|vision focuses on what is visible to the eye and applies volumetric methods that might otherwise be too tedious to use. The software does not perform any function which cannot be accomplished by a trained user utilizing manual tracing methods; the intent of the software is to save time and automate potential error prone manual tasks.
VIDA|vision has functions for loading, analyzing, and saving datasets, and will generate screen displays, computations and aggregate statistics. VIDA|vision data output may be exported in pdf format or to a csv file.
The provided document is a 510(k) Premarket Notification from the FDA to VIDA Diagnostics Inc. for their device, VIDA|vision. The document primarily focuses on demonstrating substantial equivalence to a predicate device, not on presenting detailed performance metrics and acceptance criteria from a clinical or non-clinical study for a specific AI function.
Specifically, the document states:
- "No human clinical testing was required to support a substantial equivalence finding."
- "Lung and lobe segmentation performance was tested against the predicate performance to demonstrate substantial equivalence." However, it does not provide the specific acceptance criteria or detailed results of this performance testing.
Therefore, many of the requested details about acceptance criteria and the study that proves the device meets them (such as specific performance metrics, sample sizes, expert qualifications, and ground truth methodologies) are not available in the provided text. The document focuses on regulatory compliance and comparison to a predicate, rather than a detailed performance study with quantifiable results.
Based on the information available:
1. A table of acceptance criteria and the reported device performance:
- Acceptance Criteria: Not explicitly stated in terms of specific numerical thresholds for AI performance (e.g., specific accuracy, sensitivity, specificity values). The general acceptance criterion appears to be "demonstrate substantial equivalence" to the predicate device in terms of lung and lobe segmentation performance.
- Reported Device Performance: No specific numerical performance metrics (e.g., accuracy, DICE score) are provided. It only states that "Results of testing demonstrate that the device has met all product specifications and user needs within its intended use." for lung and lobe segmentation.
2. Sample size used for the test set and the data provenance:
- Sample Size: Not specified.
- Data Provenance: Not specified (e.g., country of origin, retrospective/prospective).
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not specified.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not specified.
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 MRMC study is mentioned or summarized in the document. The statement "No human clinical testing was required" implies such a study assessing human reader improvement with AI assistance was not performed or submitted for this clearance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- It states that "Lung and lobe segmentation performance was tested against the predicate performance." This implies a standalone technical performance evaluation was done for the segmentation algorithm, but no specific metrics or methodology are provided to describe "standalone performance" in detail.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Not specified in the provided text.
8. The sample size for the training set:
- Not specified.
9. How the ground truth for the training set was established:
- Not specified.
In summary, the provided FDA 510(k) clearance letter and summary primarily address regulatory aspects, device description, and comparison to a predicate device, rather than detailed technical study results with specific performance metrics for the deep learning algorithm. The mention of "deep learning-based segmentation algorithms" is a key difference from the predicate, but the specific validation of these algorithms in terms of detailed performance data is not included in this summary document.
§ 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.