Augmenting Dementia Cognitive Assessment With Instruction-Less Eye-Tracking Tests.

Eye-tracking technology is an innovative tool that holds promise for enhancing dementia screening. In this work, we introduce a novel way of extracting salient features directly from the raw eye-tracking data of a mixed sample of dementia patients during a novel instruction-less cognitive test. Our approach is based on self-supervised representation learning where, by training initially a deep neural network to solve a pretext task using well-defined available labels (e.g. recognising distinct cognitive activities in healthy individuals), the network encodes high-level semantic information which is useful for solving other problems of interest (e.g. dementia classification). Inspired by previous work in explainable AI, we use the Layer-wise Relevance Propagation (LRP) technique to describe our network's decisions in differentiating between the distinct cognitive activities. The extent to which eye-tracking features of dementia patients deviate from healthy behaviour is then explored, followed by a comparison between self-supervised and handcrafted representations on discriminating between participants with and without dementia. Our findings not only reveal novel self-supervised learning features that are more sensitive than handcrafted features in detecting performance differences between participants with and without dementia across a variety of tasks, but also validate that instruction-less eye-tracking tests can detect oculomotor biomarkers of dementia-related cognitive dysfunction. This work highlights the contribution of self-supervised representation learning techniques in biomedical applications where the small number of patients, the non-homogenous presentations of the disease and the complexity of the setting can be a challenge using state-of-the-art feature extraction methods.

Effects of lighting variability on locomotion in posterior cortical atrophy.

INTRODUCTION: Clinical reports describe patients with Alzheimer's disease (AD) exhibiting atypical adaptive walking responses to the visual environment; however, there is limited empirical investigation of such behaviors or factors modulating their expression. We aim to evaluate effects of lighting-based interventions and clinical presentation (visual- vs memory-led) on walking function in participants with posterior cortical atrophy (PCA) and typical AD (tAD). METHODS: Participants with PCA (n = 10), tAD (n = 9), and healthy controls (n = 12) walked to visible target destinations under different lighting conditions within two pilot repeated-measures design investigations (Experiment 1: 32 trials per participant; Experiment 2: 36 trials per participant). Participants walked to destinations with the floorpath interrupted by shadows varying in spatial extent (Experiment 1: no, medium, high shadow) or with different localized parts of the environment illuminated (Experiment 2: target, middle, or distractor illuminated). The primary study outcome for both experimental tasks was completion time; secondary kinematic outcomes were proportions of steps identified as outliers (Experiment 1) and walking path directness (Experiment 2). RESULTS: In Experiment 1, PCA participants overall demonstrated modest reductions in time taken to reach destinations when walking to destinations uninterrupted by shadows compared to high shadow conditions (7.1% reduction [95% confidence interval 2.5, 11.5; P = .003]). Experiment 2 found no evidence of differences in task performance for different localized lighting conditions in PCA participants overall. Neither experiment found evidence of differences in task performance between conditions in tAD or control participants overall. Completion time in both patient groups was longer relative to controls, and longer in PCA relative to tAD groups. DISCUSSION: Findings represent a quantitative characterization of a clinical phenomenon involving patients misperceiving shadows, implicating dementia-related cortico-visual impairments. Results contribute to evidence-based design guidelines for dementia-friendly environments.