Work

Principal component analysis‐derived magnetic resonance imaging brain fluid dynamics‐related composite in the Alzheimer's spectrum

Year: 2026

Type: article

Abstract: INTRODUCTION Magnetic resonance imaging (MRI) ‐derived markers may characterize complementary structural and diffusion‐related features potentially relevant to brain fluid dynamics in Alzheimer's disease (AD). We developed an exploratory multivariate construct integrating these features and assessed clinical correlates. METHODS Fifty‐six participants who were cognitively normal (CN) or had mild cognitive impairment or AD underwent MRI. Principal component analysis (PCA) integrated three complementary MRI measures: intracranial volume‐adjusted choroid plexus volume, the diffusion tensor image analysis along the perivascular space (DTI‐ALPS) index, and neurite orientation dispersion and density imaging‐derived free‐water–related isotropic volume fraction (FW/fISO). The first component was designated the PCA‐derived MRI brain fluid dynamics‐related composite (MRI‐BFRC). RESULTS The first component explained 58.4% of the variance, with higher scores reflecting larger choroid plexus volume, lower DTI‐ALPS, and higher FW/fISO. In the primary multivariable model, clinically defined AD relative to CN and higher Fazekas total score were independently associated with higher MRI‐BFRC. MRI‐BFRC correlated strongly with ventricular volume ( r = 0.779); after adjustment for brain and ventricular volumes, the association with Fazekas score remained significant, whereas the AD–CN association was attenuated. DISCUSSION MRI‐BFRC is an exploratory multimodal construct integrating MRI features selected for their potential relevance to brain fluid dynamics. Its strong coupling with ventricular enlargement indicates sensitivity to global structural variation; persistence of the white matter hyperintensity association after structural adjustment suggests variation not fully explained by global structural measures. MRI‐BFRC may have potential as a multimodal imaging marker for evaluating brain fluid dynamics‐related alterations, although validation against independent physiological measures is required.

Source: Alzheimer s & Dementia Diagnosis Assessment & Disease Monitoring

DOI: 10.1002/dad2.70538