Aneurysm growth evaluation and detection: a computer-assisted follow-up MRA analysis

Area of Science:

• Medical Imaging
• Computational Anatomy
• Neurosurgery

Background:

• Growing intracranial aneurysms (IAs) risk rupture, necessitating accurate growth detection for treatment.
• Current methods for IA assessment may lack sensitivity in identifying subtle growth patterns.

Purpose of the Study:

• To develop and validate a computer-assisted approach for quantifying IA growth using novel imaging features.
• To assess the performance of these novel features in differentiating stable from growing IAs.

Main Methods:

• Extraction of IA shapes from baseline and follow-up angiographic scans.
• Non-rigid morphing to compute shape deformations and derive novel features: differential volume (dV), surface area (dSA), normalized median deformation path length (dMPL), and integral of cumulative deformation distances (dICDD).
• Classification of IAs by a neuroradiologist into stable (with morphology changes) or growing categories for performance evaluation.

Main Results:

• Three novel features and one cross-sectional feature showed significantly different mean values between stable and growing IA groups (p $$).
• The integral of cumulative deformation distances (dICDD) was significantly different across all three assessed IA groups.
• Novel features demonstrated higher sensitivity (0.81-0.90) for detecting growing IAs compared to cross-sectional features (0.05-0.86).

Conclusions:

• The proposed computer-assisted method and novel features show promise as sensitive, imaging-based biomarkers for IA growth detection.
• These findings could enhance clinical decision-making for managing patients with intracranial aneurysms based on follow-up imaging.