(195 days)
SwiftMR is a stand-alone software solution intended to be used for acceptance, enhancement and transfer of brain MRI images in DICOM format. It can be used for noise reduction and increasing image sharpness for non-contrast enhanced MRI images.
SwiftMR is not intended for use on mobile devices.
SwiftMR, is software used as a Medical Device (SaMD) consisting of a software algorithm that enhances images taken by MRI scanners. The device only processes DICOM images for the end User and is intended to be used by radiology technologists in an imaging center, clinic, or hospital.
The device's inputs are standard of care MRI images in DICOM format. The deep learning algorithm produces enhanced images as outputs with reduced noise and increased sharpness in DICOM format. The device applies both denoising and sharpness increase functions simultaneously.
SwiftMR provides an automatic image quality enhancement function for MR images acquired in various environments. SwiftMR can only be used for professional purposes and is not intended for use on mobile devices.
SwiftMR 's automation procedure is as follows:
- . Upload MR images that have been taken or converted to the DICOM format
- . Image quality enhancement using Deep Learning model
- . Download enhanced MR image as DICOM format
There are two deep learning algorithms that should be selected by users according to the pulse sequences. One is for the general pulse sequences and the other is for the TOF pulse sequences. The two deep learning algorithms share network architecture, input data generation method, training procedures. The only difference between the two deep learning algorithms is the input / label dataset used for training.
After integration with the facilities PACS, SwiftMR performs image processing in the background automatically. At the same time, SwiftMR allows logged-in users to use its functions and change product settings through the client application. When logged in as the System Admin, the function is available to the control automation procedure and system change settings. On the User side, the User can retrieve the results of image processing in the form of a worklist by login to the user account.
The software provides three main functions, which are image processing, quality check and progress monitoring.
The software is intended to run automatically in the background so that it does not interrupt the workflow of users. When the user executes MR scans and saves the images in PACS as he/she usually does, the newly acquired images are automatically uploaded to the server and registered in the database (DB) for image processing. Once image processing is complete, the images are sent back to PACS.
If the user wishes to monitor this automated workflow to check on the status of image processing, he/she can log in to the Client App, and notifications will be received upon each task completion. Detailed information is also available when the main worklist is opened, and MR Study and/or Series in concern is selected.
A settings menu is provided in the form of a user interface to enable users and system admin to modify software settings as required by the institution or respective user.
Here's a breakdown of the acceptance criteria and study details for SwiftMR, based on the provided FDA 510(k) summary:
1. Acceptance Criteria and Reported Device Performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Noise Reduction: Average Signal-to-Noise Ratio (SNR) of SwiftMR-processed image series is increased by 40% or more compared to the original image series. | Test passed. The average SNR of SwiftMR-processed images was increased by 40% or more. |
| Sharpness Increase: FWHM (Full Width at Half Maximum) of a selected Region of Interest (ROI) is decreased by 0.13% or more after applying SwiftMR for at least 90% of the test datasets. | Test passed. The FWHM of selected ROIs was decreased by 0.13% or more for at least 90% of the test datasets. |
2. Sample Size and Data Provenance for the Test Set
- Sample Size for Test Set: Not explicitly stated. The document mentions "retrospective clinical images" and "at least 90% of the test datasets," but the absolute number of cases or images is not provided.
- Data Provenance: Retrospective clinical images. The country of origin is not specified, but the applicant (AIRS Medical Inc.) is based in South Korea, suggesting potential involvement of data from that region.
3. Number of Experts and Qualifications for Ground Truth
- Number of Experts: Not specified.
- Qualifications of Experts: Not specified.
4. Adjudication Method for the Test Set
- Adjudication Method: Not specified. The document only states that "predetermined acceptance criteria were met," implying internal validation, but does not detail how potential disagreements in ground truth were resolved if multiple experts were used.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- MRMC Study Conducted: No, a multi-reader multi-case (MRMC) comparative effectiveness study was not reported. The performance data focuses on technical metrics (SNR and FWHM) of the algorithm's output, not on human reader improvement with or without AI assistance.
- Effect Size of Human Reader Improvement: Not applicable, as no MRMC study was performed.
6. Standalone Algorithm Performance Study
- Standalone Performance Study: Yes, a standalone (algorithm only without human-in-the-loop performance) study was conducted. The performance data presented (SNR increase and FWHM decrease) directly quantifies the technical output of the SwiftMR algorithm.
7. Type of Ground Truth Used
- Type of Ground Truth: The ground truth for the performance test appears to be based on quantitative technical metrics of image quality (SNR and FWHM) of the original images, which are then compared to the processed images. It is implicitly assumed that higher SNR and lower FWHM indicate improved image quality. It's not based on expert consensus for clinical diagnosis, pathology, or outcomes data in this summary.
8. Sample Size for the Training Set
- Sample Size for Training Set: Not explicitly stated. The document mentions "input / label dataset used for training" for two deep learning algorithms (general pulse sequences and TOF pulse sequences), but the size of these datasets is not provided.
9. How Ground Truth for the Training Set Was Established
- Ground Truth Establishment for Training Set: The document states that the two deep learning algorithms "share network architecture, input data generation method, training procedures. The only difference between the two deep learning algorithms is the input / label dataset used for training." However, it does not explicitly describe how the "label dataset" (ground truth) for training was established. It can be inferred that the labels would correspond to an idealized "enhanced" version of the input, likely generated through a process that defines optimal noise reduction and sharpness, but the specifics of this generation are not detailed.
§ 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).