An Image-Based Workflow for Objective Vessel Wall Enhancement Quantification in Intracranial Aneurysms

Area of Science:

• Medical Imaging
• Neuroradiology
• Biomarker Development

Background:

• Veno-Venous Enhancement (VWE) on contrast-enhanced MRI is a potential biomarker for intracranial aneurysm (IA) evaluation.
• Current visual inspection methods for identifying VWE in IAs are prone to errors and inconsistencies.
• An objective tool for VWE visualization and quantification is needed to enhance IA assessment reliability.

Purpose of the Study:

• To develop and validate a computational tool for objective visualization, quantification, and identification of VWE regions in IAs.
• To assess the tool's accuracy in distinguishing enhancing from non-enhancing areas within IAs.
• To determine if the enhancement area ratio (EAR) can predict high-risk intracranial aneurysms.

Main Methods:

• Co-registration of 41 3D T1-MRI and 3D TOF-MRA IA images from 38 patients.
• Normalization of contrast-enhanced MRI by pituitary stalk enhancement intensity.
• Generation of IA surface models and mapping of signal intensities; identification of optimal threshold using 30 IAs, validated on 11 IAs.

Main Results:

• An optimal normalized intensity threshold of 0.276 was identified for delineating VWE regions with 81.7% validation accuracy.
• The developed method achieved a Dice coefficient of 71.1% in overlap with neuroradiologist-marked regions.
• An enhancement area ratio (EAR) of 23% effectively discriminated high-risk aneurysms with an AUC of 0.7.

Conclusions:

• A validated pipeline for objective VWE region identification in IAs was successfully developed.
• This tool has the potential to significantly improve the reliability and consistency of IA evaluation.
• Objective VWE quantification may aid in identifying high-risk intracranial aneurysms.