Neuro Insight V1.0

Neuro Insight V1.0 is an image processing software solution composed of NeuroPro (an image processing application that computes maps and metrics from neurological MRI images) and Neuro Synchronizer (an optional interface for viewing, manipulating, and comparing medical images). It is intended to assist appropriately trained medical professionals in analyzing neurological MRI studies and does not alter original images or serve as a standalone diagnostic tool.

Both Neuro Insight V1.0 and predicate Olea Sphere V3.0 process FLAIR, T1, T2, T1g, diffusion, and DSC perfusion series and produce identical parametric map outputs (CBV, CBF, MTT, TTP, Tmax/Delay, CBV_corr, K2, tMIP). Key differences: Neuro Insight uses 3D rigid motion correction (already cleared in K223502) and a 3D deep learning brain extraction tool (achieving DICE coefficient of 0.97), while Olea Sphere uses 2D rigid correction and histogram-based thresholds. Neuro Insight also uses 3D rigid 6-dof co-registration versus Olea Sphere's DICOM tag-based alignment.

Not stated in this summary.

Neuro Insight V1.0 represents a subset of Olea Sphere V3.0, providing MR-only processing versus Olea Sphere's MR and CT capabilities. The intended use, indications, and principles of operation are substantially equivalent—both are analysis tools for clinicians reviewing neurological MRI, neither diagnoses independently, and both require clinical judgment. The minor technological differences (motion correction, brain extraction, co-registration methods) do not raise new safety or effectiveness questions because the 3D motion correction was previously cleared (K223502), the deep learning brain extraction achieved excellent validation metrics (DICE 0.97), and the user confirmation feature for adjusted parameters (arterial input function) is already present in cleared reference device Olea S.I.A Neurovascular V1.0 (K223532). Quantitative and qualitative comparisons by three board-certified neuroradiologists using 30 anonymized cases confirmed substantial equivalence of parametric map outputs, and qualitative assessment of 60 cases confirmed acceptable co-registration quality.