Swoop® Portable MR Imaging System

The Swoop® Portable MR Imaging System™ provides accessible head imaging at the patient's bedside by utilizing low-field magnetic resonance technology. Its methodology centers on deep learning algorithms for superior image reconstruction across various sequences, including newly introduced DWI capabilities, to aid in diagnosis when conventional MRI examinations are not feasible.

The Swoop® Portable MR Imaging System™ is a bedside magnetic resonance imaging device designed for imaging the internal structure of the head in adult and pediatric patients (≥ 0 years) at the point of care, particularly where full diagnostic examination is not clinically practical. It utilizes a 63.3 mT permanent magnet and a form-fitting head coil to acquire images using standard MRI contrasts. The system is operated via a commercial off-the-shelf tablet which facilitates exam setup, image viewing for quality control, and cloud storage interactions. A significant feature is its image reconstruction algorithm that employs deep learning to enhance image quality for T1W, T2W, and FLAIR sequences by reducing image noise and blurring. This version of the Swoop System incorporates software modifications that introduce a DWI pulse sequence and associated DWI image reconstruction using the Fast Iterative Shrinkage Thresholding Algorithm (FISTA). Additionally, it includes image uniformity correction and enhanced noise correction applicable to all sequence types. Non-clinical performance testing was conducted to ensure these modifications met design specifications and user requirements. This testing included software verification, software validation, and image performance assessments against NEMA and ACR standards, which confirmed the device meets all image quality criteria. Other performance aspects such as biocompatibility, cleaning/disinfection, electrical safety, and cybersecurity were addressed by leveraging existing data from a prior iteration of the device, as the new modifications did not introduce new worst-case scenarios for these aspects.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the image quality of the Swoop System's low-field (63.3 mT) MRI, particularly for the new DWI sequence, compare to conventional high-field MRI systems in detecting specific neuropathologies?; What are the quantitative metrics for the "improved image quality" and "reductions in image noise and blurring" achieved by the deep learning algorithm, especially for T1W, T2W, and FLAIR sequences?; Can the device differentiate between ischemic stroke and hemorrhagic stroke with high sensitivity and specificity using the new DWI and other available sequences at the bedside?; What is the typical acquisition time for a comprehensive head examination using the Swoop system, including T1W, T2W, FLAIR, and DWI sequences, and how does this impact its utility in acute settings?; How does the "image uniformity correction" implemented across all sequence types specifically address potential field inhomogeneities inherent in a portable, low-field magnet design?