Machine Learning and Eye Movements Give Insights into Neurodegenerative Disease Mechanisms

Abstract

Humans are a vision-dominated species; what we perceive depends on where we look. Therefore, eye movements (EMs) are essential to our interactions with the environment, and experimental findings show EMs are affected in neurodegenerative disorders (ND). This could be a reason for some cognitive and movement disorders in ND. Therefore, we aim to establish whether changes in EM-evoked responses can tell us about the progression of ND, such as Alzheimer's (AD) and Parkinson's diseases (PD), in different stages. In the present review, we have analyzed the results of psychological, neurological, and EM (saccades, antisaccades, pursuit) tests to predict disease progression with machine learning (ML) methods. Thanks to ML algorithms, from the high-dimensional parameter space, we were able to find significant EM changes related to ND symptoms that gave us insights into ND mechanisms. The predictive algorithms described use various approaches, including granular computing, Naive Bayes, Decision Trees/Tables, logistic regression, C-/Linear SVC, KNC, and Random Forest. We demonstrated that EM is a robust biomarker for assessing symptom progression in PD and AD. There are navigation problems in 3D space in both diseases. Consequently, we investigated EM experiments in the virtual space and how they may help find neurodegeneration-related brain changes, e.g., related to place or/and orientation problems. In conclusion, EM parameters with clinical symptoms are powerful precision instruments that, in addition to their potential for predictions of ND progression with the help of ML, could be used to indicate the different preclinical stages of both diseases.

Keywords: Alzheimer’s disease; Parkinson’s disease; Rough Set; eye movements; machine learning.

Figure 1

Saccades at various stages of PD. Recordings were performed with different UPDRS measured in clinical conditions under a doctor’s supervision. The red line is related to light spot movements, and the blue and green curves are related to reflexive saccades performed by two different PD patients in different stages of the disease: the blue curve denotes EM at the beginning of the disease, and the green curve at advanced PD. Notice the significant difference in response latencies [11].

Figure 2

Antisaccades (AS) at various stages of PD. The red line is related to the light spot movements, and the blue and green curves are related to reflexive AS performed by two different PD patients in different stages of the disease: the blue curve denotes EM at the beginning of the disease, and the green curve in the advanced PD stage. Notice the significant difference not only in response latencies, but also in AS speed, and that a more advanced PD patient started with a saccade that changed to AS [21].

Figure 3

A comparison of the average results +/− SD of the AS latencies, obtained in a group of 27 age-matched normal controls (healthy people), 10 AD subjects, and 12 patients with PD. The averages of the control group and AD patients are derived from the research results obtained from [43] and the average of the PD patients from the results obtained from [11].

Figure 4

The prototype of an application testing the recognition of numbers and letters in the correct order implemented in a VR environment on an Oculus Go device. Trail B is generally thought to be sensitive to executive functioning since it requires a wide range of skills to complete. In 2D, it engages attention, memory, visual screening abilities, motor functioning, and cognitive processes, and it is likely to be even more difficult in 3D. Squares in green are already marked, and in white will be chosen in the proper order by the pointer in pink.