FDA Grants De Novo Classification to neuropacs, an AI MRI Diagnostic Aid

Key points:

• FDA granted De Novo classification to neuropacs™, an AI-based MRI diagnostic aid for Parkinsonian syndromes
• The software analyzes diffusion MRI to help differentiate MSA and PSP from Parkinson’s disease
• It is intended as a supplemental tool for clinicians, not a standalone diagnostic
• The classification establishes a new regulatory category for Parkinsonian diagnostic aids
• Supporting evidence includes a multicenter study evaluating imaging-based differentiation of these conditions

The U.S. Food and Drug Administration has granted De Novo classification to neuropacs™, an artificial intelligence–based MRI software designed to assist clinicians in evaluating Parkinsonian syndromes, according to a company announcement.

The decision establishes a new regulatory category—“Parkinsonian syndrome diagnostic aid”—with neuropacs™ described as the first device cleared under this classification. The software is intended to provide additional information to support clinical assessment, rather than serve as a standalone diagnostic tool.

neuropacs™ analyzes diffusion MRI scans using machine learning to identify patterns of neurodegeneration associated with conditions such as multiple system atrophy (MSA) and progressive supranuclear palsy (PSP). The resulting report is designed to help clinicians differentiate these disorders from Parkinson’s disease (PD), which can present with overlapping symptoms, particularly in early stages.

The software is compatible with standard diffusion MRI sequences acquired on widely used 3T MRI systems and is delivered through a cloud-based workflow integrated into existing imaging practices. According to the company, the technology is supported by more than 15 years of research and development and has been trained on over 1,000 imaging datasets.

Evidence supporting the approach includes a prospective multicenter study published in JAMA Neurology, conducted across 21 movement disorder centers within the Parkinson Study Group. The study evaluated the performance of diffusion MRI–based imaging biomarkers in distinguishing atypical Parkinsonian syndromes from Parkinson’s disease under defined conditions.

Parkinsonian syndromes are often difficult to distinguish clinically, particularly early in disease progression, contributing to diagnostic uncertainty. The company positions neuropacs™ as a tool to provide objective, imaging-derived insights to support decision-making alongside standard clinical evaluation.

The De Novo classification enables commercialization of the technology and may facilitate broader adoption of AI-based imaging tools in neurology. The company also noted that recent funding will support its transition from research to commercial deployment and expansion of clinical and trial-related applications.