G-ALPS: An index for evaluating cognitive decline and aging using diffusion tensor imaging

The glymphatic system dysfunction is associated with cognitive decline in neurodegenerative diseases such as Alzheimer's disease (AD).

We introduce the G-Along Perivascular Space (G-ALPS) index, an optimized version of the ALPS index derived using Genetic Programming, and use it to analyze 217 diffusion tensor imaging (DTI) samples.

Compared to the ALPS, the proposed G-ALPS index shows a stronger correlation with cognitive measures, including Mini-Mental State Examination (MMSE, 2.78% improvement), Clinical Dementia Rating (CDR, 5.13% improvement), and Functional Activities Questionnaire (FAQ, 10% improvement), as demonstrated by the analysis of fiber diffusivities in DTI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. Furthermore, the G-ALPS exhibits enhanced sensitivity in identifying the effects of aging (94.81% improvement in AD individuals, 105% improvement in patients with mild cognitive impairment [MCI], and 81.25% improvement in normal controls) and sleep-related disorders (21.27% improvement in correlation with MMSE, and 2.53% improvement in correlation with Pittsburgh Sleep Quality Index [PSQI]) using the Human Connectome Project (HCP) dataset.

Our results suggest that the G-ALPS index may be an indirect metric for assessing the glymphatic system's function or dysfunction.