Multifractal MRI Analysis Reveals Brain Structure Loss in Aging and Dementia

A team led by Marta Lotka at the Jagiellonian University introduced Multifractal Space-filling Curve Analysis (MFSCA), a novel method that projects multidimensional MRI data onto a one-dimensional signal using a fractal space-filling curve. This preserves both local and long-range structural correlations, allowing standard multifractal algorithms (like the MF-DFA) to quantify the complexity of brain spatial organization. The method was tested on artificial fractal structures and real MRI data from healthy subjects and Alzheimer's patients at various stages of dementia.

The analysis revealed that brain spatial organization progressively weakens with age and dementia: healthy young brains exhibit strong multifractality (rich, heterogeneous structure), which transitions to monofractality (simpler, homogeneous structure) in elderly controls and—critically—in early dementia and mild cognitive impairment patients of the same age. This suggests that loss of structural complexity is a marker of both normal aging and pathological neurodegeneration. The findings demonstrate that MFSCA can provide a quantitative, non-invasive biomarker for dementia development using standard MRI, potentially enabling earlier detection and monitoring of disease progression.