Diagnostic Performance of Automated MRI Volumetry by icobrain dm for Alzheimer's Disease in a Clinical Setting: A REMEMBER Study

Area of Science:

• Neurology
• Radiology
• Medical Imaging

Background:

• Magnetic resonance imaging (MRI) is crucial for diagnosing neurodegenerative diseases.
• Automated brain volumetry software, icobrain dm, has demonstrated potential in research settings for differentiating Alzheimer's disease dementia (ADD) from healthy controls (HC).

Purpose of the Study:

• To evaluate the diagnostic value of icobrain dm in routine clinical practice for ADD diagnosis.
• To compare icobrain dm's performance against FreeSurfer, a widely used software.
• To investigate the contribution of combined brain volumes to ADD diagnosis.

Main Methods:

• Retrospective, multi-center study (REMEMBER) involving 820 participants (HC, SCD, MCI, ADD).
• Automated volumetric analysis of brain structures using icobrain dm (v.4.4.0) and FreeSurfer (v.6.0) on clinical brain MRI T1w images.
• Receiver operating characteristic (ROC) analysis and stepwise backward regression were used for comparisons and diagnostic accuracy assessment.

Main Results:

• icobrain dm significantly outperformed FreeSurfer in processing time and robustness.
• icobrain dm demonstrated superior diagnostic performance for whole brain, hippocampal volumes, and lateral ventricles between HC and ADD patients.
• Stepwise backward regression indicated improved diagnostic accuracy when combining brain volumes, particularly for distinguishing HC from ADD (AUC=0.914).

Conclusions:

• Automated volumetry, specifically icobrain dm, is valuable for diagnosing ADD in routine clinical practice.
• The combination of brain volumes derived from automated analysis enhances diagnostic accuracy for ADD.
• Real-world clinical imaging data supports the utility of icobrain dm in diagnosing Alzheimer's disease.