Search Results
Found 1 results
510(k) Data Aggregation
(84 days)
MFI-Cardiac
The MFI-Cardiac application is intended to provide processing of rest and stress myocardial perfusion positron emission tomography (PET) images with approved radiotracers that have been acquired with a singlescan or fast back-to-back scans without waiting for radiotracer decay between scans. The application models measurements of tracer uptake over time to aid in the processing of the images. Co-registration or fusion of volumetric data is provided as a quality control. The processed output images are reviewed on other devices by qualified medical professionals trained in the use of medical imaging devices.
MFI-Cardiac is a software medical device that facilitates the rapid acquisition and processing of rest and stress myocardial perfusion images (MPI) with positron emission tomography (PET) and approved radiotracers. The device processes rest+stress cardiac PET images that have been rapidly acquired using either a single scan with two administrations of the tracer (one at rest, the other at stress) or separate "back-to-back" rest and stress scans. The rapidly acquired rest and stress cardiac PET images are provided to MFI—Cardiac via digital image transfer in DICOM format.
In use, the device first performs a series of quality assurance tests to ensure data integrity and validity. The images are then processed to correct for residual activity from the 1st tracer administration (administered at rest) that affects the image obtained after the 2nd tracer injection (administered at stress). The device then outputs separated and corrected DICOM images at rest and stress; these images are offloaded to other devices for display and review by qualified healthcare professionals.
Here's a detailed breakdown of the acceptance criteria and study information for the MFI-Cardiac device, based on the provided FDA 510(k) summary:
MFI-Cardiac Acceptance Criteria and Study Information
The MFI-Cardiac device underwent performance evaluations to demonstrate its safety and effectiveness, including simulation studies and evaluations using patient datasets. The device is classified as having a "moderate" level of concern, meaning a software failure could directly result in minor patient injury.
1. Table of Acceptance Criteria and Reported Device Performance
The FDA 510(k) summary states that "Three performance evaluation studies were completed with all acceptance criteria met." However, the document does not explicitly list the quantitative acceptance criteria for each study. It describes the purpose of each study. Therefore, the "Reported Device Performance" column reflects that the studies were successful in meeting their unstated criteria.
| Acceptance Criteria (Inferred from Study Purpose) | Reported Device Performance |
|---|---|
| Simulation Studies: Evaluate device performance across the full range of rest and stress myocardial blood flows anticipated clinically. (Implies accuracy and reliability of measurements under various physiological conditions). | All acceptance criteria met. (Device performed successfully across the full range of simulated blood flows.) |
| Patient Dataset with Synthesized Rest+Stress Images (Performance Evaluation): Evaluate the performance of MFI-Cardiac for producing output (corrected) rest and stress images that are equivalent to rest and stress images obtained from separate scans. (Implies accuracy of correction algorithm). | All acceptance criteria met. (Corrected images were equivalent to those from separate scans.) |
| Patient Dataset with Synthesized Rest+Stress Images (Co-registration Evaluation): Evaluate rest-stress image co-registrations performed by MFI-Cardiac for ensuring that the estimate for residual rest tracer activity is aligned with the stress-phase images before the correction is applied. (Implies accuracy of co-registration). | All acceptance criteria met. (Co-registration ensured proper alignment for correction.) |
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Sample Size: The document does not specify the exact number of cases or images used in the "patient dataset" for the performance evaluation and co-registration evaluation studies. It refers to "patient dataset with synthesized rest+stress images" in the plural, implying more than one, but no specific count.
- Data Provenance: Not explicitly stated. The document mentions "patient dataset," but does not specify the country of origin or whether the data was retrospective or prospective. The term "synthesized rest+stress images" suggests that existing patient data might have been manipulated or combined to create the test cases.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The document does not provide information on the number or qualifications of experts used to establish ground truth for the test set.
4. Adjudication Method for the Test Set
The document does not provide information on any adjudication method used for the test set.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance
No, an MRMC comparative effectiveness study involving human readers with and without AI assistance was not explicitly mentioned or described in the provided text. The device is for image processing and output, which are then reviewed by qualified medical professionals. The focus of the studies described is on the algorithm's performance in processing and correcting images.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
Yes, the performance evaluation studies described are essentially standalone (algorithm only) performance assessments.
- "Simulation studies to evaluate device performance across the full range of rest and stress myocardial blood flows..."
- "Performance evaluation using patient dataset with synthesized rest+stress images to evaluate the performance of MFI-Cardiac for producing output (corrected) rest and stress images..."
- "Evaluation of rest-stress image co-registrations performed by MFI-Cardiac for ensuring that the estimate for residual rest tracer activity is aligned with the stress-phase images before the correction is applied."
These studies assessed the device's ability to process, correct, and align images autonomously, without direct human intervention in the real-time processing loop. The output images are then reviewed by humans, but the studies described focus on the algorithm's direct output.
7. The Type of Ground Truth Used
The ground truth appears to be established through simulation and comparison to "separate scans."
- For the simulation studies, the ground truth would be the known, controlled parameters of the simulated myocardial blood flows.
- For the performance evaluation using patient dataset with synthesized rest+stress images, the ground truth for the "corrected" images was implicitly the "rest and stress images obtained from separate scans." This suggests a comparison against an existing, established method or "gold standard" for acquiring uncorrected images.
- For the co-registration evaluation, the ground truth would involve the accurate anatomical alignment of different image phases, likely assessed computationally against ideal alignment.
8. The Sample Size for the Training Set
The document does not provide any information regarding the sample size used for the training set.
9. How the Ground Truth for the Training Set Was Established
The document does not provide any information on how the ground truth for the training set, if any, was established. Given the nature of the device (processing and correction based on physical models rather than deep learning for classification), it's possible that traditional algorithmic development and tuning did not involve a distinct "training set" in the same way a deep learning model would.
Ask a specific question about this device
Page 1 of 1