EFAI Neurosuite CT Midline Shift Assessment System (MLS-CT-100)

EFAI MLSCT is a software tool that analyzes non-contrast head CT scans using deep learning to automatically identify cases with suspected midline shift (MLS). It provides case-level notifications to radiologists through PACS/workstation to help prioritize clinical review, without marking specific image locations or replacing standard clinical assessment.

Both the proposed device and predicate use artificial intelligence algorithms with image databases to analyze non-contrast head CTs. Key differences: proposed device detects only midline shift with no user configuration options, whereas predicate detects four findings (ICH, mass effect, MLS, cranial fracture) with user controls. Proposed device does not provide preview images; predicate does. Proposed device deploys only on-premise; predicate can deploy on-premise or cloud with HIPAA-compliant interfaces.

IEC 62304:2006/A1:2016 (Medical device software – Software life cycle processes); ISO 14971:2019 (Application of risk management to medical devices); FDA guidance documents on Device Software Functions and Cybersecurity in Medical Devices.

The proposed device and predicate both provide case-level triage notifications for non-contrast head CT using AI algorithms to flag suspected pathologies without altering original images or providing diagnostic segmentation. Although the proposed device focuses exclusively on midline shift without user configuration and the predicate covers four findings with configurable triage, this narrower scope does not raise new safety or effectiveness questions because both devices operate identically in workflow—as passive notifications alongside standard clinical interpretation. The proposed device's clinical validation demonstrated sensitivity 0.961 and specificity 0.955 (substantially equivalent to predicate's performance), and both use identical underlying technology. The on-premise-only deployment of the proposed device poses no new risks compared to the predicate's cloud option, since both present the same notifications to clinicians who retain full responsibility for image review.