Machine learning detects altered spatial navigation features in outdoor behaviour of Alzheimer’s disease patients

Abstract

Impairment of navigation is one of the earliest symptoms of Alzheimer’s disease (AD), but to date studies have involved proxy tests of navigation rather than studies of real life behaviour. Here we use GPS tracking to measure ecological outdoor behaviour in AD. The aim was to use data-driven machine learning approaches to explore spatial metrics within real life navigational traces that discriminate AD patients from controls. 15 AD patients and 18 controls underwent tracking of their outdoor navigation over two weeks. Three kinds of spatiotemporal features of segments were extracted, characterising the mobility domain (entropy, segment similarity, distance from home), spatial shape (total turning angle, segment complexity), and temporal characteristics (stop duration). Patients significantly differed from controls on entropy (p-value 0.008), segment similarity (p-value $${10}^{-7}$$ 10 - 7 ), and distance from home (p-value $${10}^{-14}$$ 10 - 14 ). Graph-based analyses yielded preliminary data indicating that topological features assessing the connectivity of visited locations may also differentiate patients from controls. In conclusion, our results show that specific outdoor navigation features discriminate AD patients from controls, which has significant implication for future AD diagnostics, outcome measures and interventions. Furthermore, this work illustrates how wearables-based sensing of everyday behaviour may be used to deliver ecologically-valid digital biomarkers of AD pathophysiology.

Exploration patterns shape cognitive map learning

“…Further, humans with higher roaming entropy in their day-to-day GPS traces have stronger hippocampal-striatal connectivity relative to those who explore less (Heller et al, 2020). A recent study also found that patients with Alzheimer's Disease show significantly reduced entropy compared to healthy older adults (Ghosh et al, 2022). To our knowledge, the inter-individual variability in this measure of exploration has not been linked to the subsequent structure of cognitive maps.…”

Geometry of navigation identifies genetic-risk and clinical Alzheimer’s disease

“…A more careful analysis involving a larger dataset of clinically labelled navigations is necessary to achieve this. The comprehensive behavioural traits captured by our metrics, combined with their efficiency in an on-line context, make them particularly promising to study ethological data sets, for instance based on GPS, where they could be used to quantify navigation of birds, insects, and people (Ghosh et al 2022, Torus et al 2017, Thums et al 2018.…”

The role of geographic information system and global positioning system in dementia care and research: a scoping review

“…Eighteen studies were identified in this category, and all applied GPS as a space-based technology to gather tracking data in exploring patients' behavior patterns. In most studies, analysis and interpretation of spatial GPS data were used to assess patients' out-of-home behavior [20][21][22][23], mobility patterns [52,53,65,[114][115][116][117], life-space metrics [118,119], and driving behavior [24,25]. In one study, GPS data were used to propose a Bayesian classifier model to estimate the probability of wandering [120].…”

“…GPS was originally intended for use in the military but then widely used in healthcare, especially for mental health disorders, to monitor, follow, track and manage the care process of patients [18,19]. GPS assesses and analyses patients' out-of-home and driving behavior [20][21][22][23][24][25]. Out-of-home behavior patterns can predict cognitive impairment disorders [21].…”