Abstract:
This paper presents a low-cost eye-tracker aimed at carrying out tests based on a Visual Paired Comparison protocol for the early detection of Mild Cognitive Impairment. The proposed eye-tracking system is based on machine learning algorithms, a standard webcam, and two personal computers that constitute, respectively, the ”Measurement Sub-System” performing the test on the patients and the ”Test Management Sub-System” used by medical staff for configuring the test protocol, recording the patient data, monitoring the test, and storing the test results. The system also integrates a stress estimator based on the measurement of heart rate variability obtained with photoplethysmography.
Index Terms—Eye-tracker, Alzheimer’s disease, Mild Cognitive Impairment, Early Detection, Raspberry, Webcam, Heart Rate Variability, Photoplethysmography, Python, Neural Networks.
Introduction:
Eye-trackers are devices designed to measure the direction of the gaze (where one is looking) and/or both eye movements related to the head. Nowadays, they are widely used in many fields, ranging from neuroscience, marketing, product design, and human-machine interaction. There are several methods for measuring eye movements, including search coils and electrooculograms, but currently, the most popular eye-trackers use video images from which the eye position is extracted. Some cognitive impairments, such as Alzheimer’s Disease (AD), are linked with eye movement-related disorders, while others indirectly relate to the gaze dynamics. Several investigations have been conducted in the last few years aimed at applying eye-tracking technology to the detection of AD.
The main objective of medical research on AD is to diagnose the disease as soon as possible to maximize the effectiveness of therapies. It is even possible to identify this disorder before symptoms appear by detecting Mild Cognitive Impairment (MCI). Patients with symptoms of MCI have a very high risk of developing AD, with a conversion rate of 6% to 25% per year. Although many patients with MCI tend to become stable after some time, more than half of them regress into dementia within five years, which outlines MCI as an excellent alarm bell that anticipates AD development.
Even if MCI does not markedly impair the daily activities of a person, some symptoms consist in difficulties in carrying out complex operations that previously took place without problems (such as preparing a meal). It is therefore extremely important to identify symptoms of MCI from its early stages. Diagnostic biomarkers able to identify the pathology associated with this disorder include the detection of altered levels of tau and amyloid in cerebrospinal fluid, the use of structural Magnetic Resonance Imaging (MRI) to identify disease-specific patterns of regional atrophy and MRI T1ρ to detect disease-related macro-molecular protein aggregation, and the direct imaging of amyloid deposits using positron emission tomography and single photon emission computerized tomography.
As it is not possible to foreseen a wide diffusion of all those diagnostic tests to the entire adult population, given their high costs and degree of invasiveness, eye-tracking-based tests are very promising in order to obtain reliable, low-cost, and non-invasive biomarkers for the early detection of MCI. In fact, there are at least two types of tests based on eye-tracking that allow us to obtain biomarkers for MCI: