Every human brain is covered with folds, much like a fingerprint: each with its own unique pattern. In dementia, this pattern changes, but for two similar diseases, Alzheimer's and dementia with Lewy bodies, the changes are nearly indistinguishable to the eye. Doctors need a reliable way to separate them, because the treatments differ.
The scientists focused on special points where three folds converge, such junctions on the brain's surface. Connecting these points with lines yields a network resembling a road map. But each person's junctions are arranged differently, making direct comparisons difficult: the maps differ in size and shape.
The solution came unexpectedly: they set a random walker loose on the network, choosing a road at random at each junction. After many steps, they collect statistics — how often the walker visited various locations. This 'walk portrait' is independent of how the junctions are numbered, so networks from different people can be compared fairly.
The method was tested on a large group of patients with both diagnoses. The new fold-based network proved more accurate than previous ones that relied on an averaged brain atlas. This means the individual pattern of folds holds important information about the disease that the template erases.
Folds appear before birth: the fetal brain folds like a sheet of paper to fit more cells into the small volume of the skull.