(84 days)
XENOVIEW VDP is image processing software that analyzes a pulmonary hyperpolarized 129-Xe MR image and a proton chest MR image to provide visualization of lung ventilation in adults and pediatric patients aged 12 years and older.
XENOVIEW VDP is an image analysis platform that quantifies normalized xenon intensity of a ventilated space using a pulmonary hyperpolarized 129-Xe ventilation MR image and accompanying proton chest MR image. The XENOVIEW VDP image analysis process includes loading and viewing images, image registration and seqmentation, normalization and statistical analysis of 129Xe signal intensity distribution within the ventilation scan, and ultimately reporting the fraction of ventilated lunq volume as a percentage of thoracic cavity volume. This software will be used by clinicians to assist in the interpretation and numerical classification of hyperpolarized 129-Xe ventilation MR images. The HP 129Xe ventilation MR images are generated using an MRI scanner and appropriate RF chest coil with a patient that has inhaled XENOVIEW (xenon Xe 129 hyperpolarized). The software provides a user-friendly interface and simple workflow that helps quide the user through the image analysis process, including the loading of images, registration of the anatomical image sets to the HP 129Xe image sets, segmentation of the lung, and automated classification of normalized ventilation distribution into multiple intensity levels via analysis of hyperpolarized 129Xe signal intensity within the segmented lung volume. The results of the image analysis are output as medical images of the classified ventilation, a summary report, and data files containing quantitative statistical analysis results.
The provided document is a 510(k) Premarket Notification from the FDA, seeking substantial equivalence for the XENOVIEW VDP device. While it summarizes non-clinical testing for software verification and validation, it explicitly states that clinical studies were not required. Therefore, the document does not contain the detailed clinical study information requested to describe acceptance criteria and proof of performance through clinical trials.
The document focuses on demonstrating substantial equivalence to a predicate device (GE Medical Systems Thoracic VCAR K103480) based on technological characteristics and non-clinical performance.
Here's an analysis based on the information provided in the document, and what is missing:
The document states:
- "Summary of Clinical Testing: XENOVIEW VDP did not require clinical studies to support substantial equivalence."
This immediately indicates that information regarding a clinical study proving the device meets specific acceptance criteria as you've requested (e.g., sample size, expert consensus, MRMC studies, specific performance metrics against ground truth from a clinical study) will not be present.
Therefore, for aspects relating to clinical study data, such as a table of acceptance criteria met by a clinical study, sample sizes, expert ground truth, adjudication methods, MRMC studies, or standalone performance from a clinical study, the answer is that this information is not available in the provided text because clinical studies were not required for this 510(k) submission.
The "acceptance criteria" discussed in this document pertain to the software verification and validation testing and demonstration of substantial equivalence to a predicate device, not performance within a clinical setting.
However, I can still address the aspects for which information is provided or implied by the nature of a 510(k) submission not requiring clinical trials:
1. A table of acceptance criteria and the reported device performance:
- Acceptance Criteria (from a 510(k) perspective for software verification/validation): The document relies on "software verification and validation testing per FDA's guidance 'Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices' (May 11, 2005)." This guidance typically involves demonstrating that the software functions as intended and meets its specifications, including requirements for accuracy, reliability, and security of its image processing functions. Specific quantitative performance metrics for disease detection accuracy against clinical ground truth are not provided because a clinical study was not conducted or required.
- Reported Device Performance: The document states, "The safety and performance of the XENOVIEW VDP software has been evaluated and verified in accordance with software specifications through software verification and validation testing." No specific numerical performance metrics (e.g., sensitivity, specificity, accuracy for a clinical outcome) are reported as this was not a clinical trial. The performance reported here refers to the successful completion of engineering validation tests.
2. Sample size used for the test set and the data provenance:
- Clinical Test Set: Not applicable/not provided, as clinical studies were not required.
- Software Verification/Validation Test Set: The document does not specify the sample size or provenance of data used for the software verification and validation tests. This type of detail is typically found in the full 510(k) submission, not the public summary letter. These tests would likely use a combination of synthetic, historical, or internally generated data to test functionality, robustness, and accuracy of image processing steps (e.g., registration, segmentation, quantification of xenon intensity). Provenance might be internal datasets or publicly available phantom/synthetic data.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Clinical Test Set: Not applicable/not provided, as clinical studies were not required. For software verification, ground truth would likely be established by engineering specifications, known inputs, or validated reference images, rather than clinical experts.
4. Adjudication method for the test set:
- Clinical Test Set: Not applicable/not provided.
5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done:
- No. The document explicitly states that "XENOVIEW VDP did not require clinical studies to support substantial equivalence." An MRMC study is a type of clinical study.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Clinical Standalone Performance: Not applicable/not provided. The document describes the device as "image processing software that analyzes a pulmonary hyperpolarized 129-Xe MR image and a proton chest MR image to provide visualization of lung ventilation." Its function is to assist clinicians, not to make a diagnosis independently. While its internal algorithms perform computations "standalone," the performance evaluation described here (non-clinical) does not speak to its standalone diagnostic performance in a clinical context.
7. The type of ground truth used:
- Clinical Trial Ground Truth: Not applicable/not provided.
- Software Verification/Validation Ground Truth: For the non-clinical software verification and validation, ground truth would be based on "software specifications." This means the expected outcome of processing an image (e.g., correct segmentation mask, accurate numerical quantification of xenon intensity per voxel/region) is predefined based on the design requirements and validated algorithms. This is not derived from expert clinical consensus, pathology, or patient outcomes data as it would be in a clinical study.
8. The sample size for the training set:
- Not provided. The document describes a medical device (software for image processing), focusing on its functionality and equivalence, not an AI/ML model for which a distinct "training set" would typically be discussed in a submission. While the software likely uses algorithms that were developed or "trained" at some point (e.g., for image registration or segmentation), the FDA submission for this device (XENOVIEW VDP) does not provide details on a specific training dataset from a machine learning perspective. It's listed as a "Medical Imaging Software" (K223071), not explicitly as an AI/ML device in the way recent submissions are categorized.
9. How the ground truth for the training set was established:
- Not provided. As above, the summary does not detail the training or ground truth for internal algorithms, focusing instead on the external validation of the software's functionality and performance against its specifications.
In summary, the provided FDA 510(k) letter for XENOVIEW VDP establishes substantial equivalence based on non-clinical software verification and validation testing in accordance with FDA guidance, and a comparison of technological characteristics to a predicate device. It explicitly states that clinical studies were not required or performed for this submission. Therefore, the detailed clinical study-related information requested (e.g., acceptance criteria confirmed by clinical data, sample sizes from clinical trials, expert ground truth for clinical cases, MRMC studies) is not present in the document.
§ 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).