SwiftSight-Brain

SwiftSight-Brain is an automated MRI image analysis software that automatically segments, labels, and measures volumes of brain structures and lesions from 3D T1-weighted and T2-weighted FLAIR MRI scans. It generates morphometric reports comparing measured volumes to age and gender-matched reference percentile data and is intended to support clinicians in assessment of structural MRIs.

SwiftSight-Brain uses deep learning for automated segmentation and quantification of brain structures, operates as software on Windows-based off-the-shelf hardware, processes DICOM-format MRI input through an automated pipeline, and outputs segmented color overlays with morphometric reports for PACS display. It includes automated quality control functions such as tissue contrast checks and scan protocol verification, matching the predicate device's core functionality.

Not stated in this summary.

The device is substantially equivalent because it shares identical intended use, target anatomical site (brain), processing architecture, data source (3D T1 and FLAIR MRI), and output format (segmented images and morphometric reports) with the predicate Neurophet AQUA V3.1. Performance validation demonstrates equivalent accuracy (Dice coefficients >80% for subcortical structures, >75% for cortical) and reproducibility (≤5% volume differences) compared to expert manual segmentation. Minor terminology differences in quality control functions and user qualifications do not introduce new safety risks and do not affect the device's safety or effectiveness.