Viz Subdural+, Viz SUBDURAL PLUS

Viz Subdural+ is software that automatically analyzes non-contrast head CT scans to identify, label, and measure collections of fluid in the subdural space (between the brain and skull). It reports the volume and maximum width of these collections and measures midline shift. The output is reviewed by qualified physicians alongside original images to assist in clinical assessment.

Both Viz Subdural+ and its predicate (Viz HDS) use locked AI/machine learning algorithms with deep-learning convolutional neural networks to process single-timepoint NCCT images. Both automatically receive, assess applicability of, and process imaging using similar pipeline architectures. Both output results in DICOM format to a PACS server. Key difference: Subdural+ measures subdural collections and widest width, while HDS measures intracranial hyperdensities and lateral ventricles, though both measure midline shift.

Not stated in this summary.

Subdural+ is substantially equivalent because it uses the same fundamental technology (locked AI/ML algorithm, similar neural network architecture, single timepoint NCCT input, DICOM output) as the predicate HDS and follows the same workflow and principles of operation. Although Subdural+ measures different anatomical structures (subdural collections vs. hyperdensities/ventricles), performance testing demonstrated acceptable measurement accuracy (MAE 7.53 mL for volume, 1.77 mm for width, 1.1 mm for midline shift). The differences in measured structures and overlay visualization do not raise new safety or efficacy questions because they represent standard radiological information already available in original images.