Annalise Enterprise

Annalise Enterprise is an AI-based software tool that analyzes non-contrast brain CT scans to identify suspected acute ischemic infarcts and prioritizes them in a clinical worklist. It interfaces with PACS/RIS systems to provide notifications to trained clinicians, enabling earlier evaluation of high-risk patients without replacing standard diagnostic interpretation or advanced imaging.

The subject device detects acute infarct across multiple cerebral territories (ACA, MCA, PCA, cerebellum, basilar, watershed) on non-contrast CT, whereas the predicate (Rapid NCCT Stroke) detects only large vessel occlusion in ICA and MCA-M1. Both use AI algorithms and operate in parallel to clinical workflow; performance differs with subject achieving 84.5–89.2% sensitivity and 84.1–93.1% specificity across operating points, compared to predicate's 63.5% sensitivity and 95.1% specificity.

ISO 13485 (QMS), ISO 14971 (risk management), IEC 62304 (software lifecycle), IEC 62366-1 (usability), AAMI TIR 57 (cybersecurity), ISO/IEC 27001 (information security), IEC 82304-1 (health software), DICOM (imaging standard), and FDA guidance on software submissions (2023) and CAD devices (2022).

Although the subject device detects acute infarct across broader territories than the predicate's LVO-only focus, both employ similar AI principles, operate parallel to standard care, provide worklist prioritization via PACS integration, and achieve clinically effective triage. Standalone performance testing demonstrates high sensitivity/specificity exceeding 80% at multiple operating points, comparable triage turnaround time (~82 seconds), and no new safety or effectiveness questions. The differences raise only technological—not clinical—questions and are supported by robust validation data with diverse patient demographics and equipment manufacturers.