The Potential of Naturalistic Eye Movement Tasks in the Diagnosis of Alzheimer's Disease: A Review

EM-COGLOAD: An investigation into age and cognitive load detection using eye tracking and deep learning

Alzheimer's Disease is the most prevalent neurodegenerative disease, and is a leading cause of disability among the elderly. Eye movement behaviour demonstrates potential as a non-invasive biomarker for Alzheimer's Disease, with changes detectable at an early stage after initial onset. This paper introduces a new publicly available dataset: EM-COGLOAD (available at https://osf.io/zjtdq/, DOI: 10.17605/OSF.IO/ZJTDQ). A dual-task paradigm was used to create effects of declined cognitive performance in 75 healthy adults as they carried out visual tracking tasks. Their eye movement was recorded, and time series classification of the extracted eye movement traces was explored using a range of deep learning techniques. The results of this showed that convolutional neural networks were able to achieve an accuracy of 87.5% when distinguishing between eye movement under low and high cognitive load, and 76% when distinguishing between the oldest and youngest age groups.

Oculomotor atypicalities in motor neurone disease: a systematic review

Introduction

Cognitive dysfunction is commonplace in Motor Neurone Disease (MND). However, due to the prominent motor symptoms in MND, assessing patients' cognitive function through traditional cognitive assessments, which oftentimes require motoric responses, may become increasingly challenging as the disease progresses. Oculomotor pathways are apparently resistant to pathological degeneration in MND. As such, abnormalities in oculomotor functions, largely driven by cognitive processes such as saccades and smooth pursuit eye movement, may be reflective of frontotemporal cognitive deficits in MND. Thus, saccadic and smooth pursuit eye movements may prove to be ideal mechanistic markers of cognitive function in MND.

Methods

To ascertain the utility of saccadic and smooth pursuit eye movements as markers of cognitive function in MND, this review summarizes the literature concerning saccadic and smooth pursuit eye movement task performance in people with MND.

Results and discussion

Of the 22 studies identified, noticeable patterns suggest that people with MND can be differentiated from controls based on antisaccade and smooth pursuit task performance, and thus the antisaccade task and smooth pursuit task may be potential candidates for markers of cognition in MND. However, further studies which ascertain the concordance between eye tracking measures and traditional measures of cognition are required before this assumption is extrapolated, and clinical recommendations are made.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=376620, identifier CRD42023376620.

Deep Learning-Based Eye-Tracking Analysis for Diagnosis of Alzheimer's Disease Using 3D Comprehensive Visual Stimuli

Alzheimer's Disease (AD) is a neurodegenerative disorder that causes a continuous decline in cognitive functions and eventually results in death. An early AD diagnosis is important for taking active measures to slow its deterioration. Traditional diagnoses are usually based on clinical experience, which is limited by several realistic factors. In this paper, we focus on exploiting deep learning techniques to diagnose AD based on eye-tracking behaviors. Visual attention, as a typical eye-tracking behavior, is of great clinical value in detecting cognitive abnormalities in AD patients. To better analyze the differences in visual attention between AD patients and normals, we first conducted a 3D comprehensive visual task on a noninvasive eye-tracking system to collect visual attention heatmaps. Then a multilayered comparison convolutional neural network (MC-CNN) is proposed to distinguish the visual attention differences between AD patients and normals. In MC-CNN, the multilayered feature representations of heatmaps were obtained by hierarchical residual blocks to better encode eye movement behaviors, which were further integrated into a distance vector to benefit the comprehensive visual task. From evaluation, MC-CNN can distinguish AD patients from normals with 0.84 accuracy, 0.86 recall, 0.82 precision, 0.83 F1-score and 0.90 area under the curve (AUC). The above results demonstrate the effectiveness of the proposed MC-CNN in AD diagnosis based on the comprehensive 3D visual task.

A detection model of cognitive impairment via the integrated gait and eye movement analysis from a large Chinese community cohort

INTRODUCTION

Whether the integration of eye-tracking, gait, and corresponding dual-task analysis can distinguish cognitive impairment (CI) patients from controls remains unclear.

METHODS

One thousand four hundred eighty-one participants, including 724 CI and 757 controls, were enrolled in this study. Eye movement and gait, combined with dual-task patterns, were measured. The LightGBM machine learning models were constructed.

RESULTS

A total of 105 gait and eye-tracking features were extracted. Forty-six parameters, including 32 gait and 14 eye-tracking features, showed significant differences between two groups (P < 0.05). Of these, the Gait_3Back-TurnTime and Dual-task cost-TurnTime patterns were significantly correlated with plasma phosphorylated tau 181 (p-tau181) level. A model based on dual-task gait, dual-task smooth pursuit, prosaccade, and anti-saccade achieved the best area under the receiver operating characteristics curve (AUC) of 0.987 for CI detection, while combined with p-tau181, the model discriminated mild cognitive impairment from controls with an AUC of 0.824.

DISCUSSION

Combining dual-task gait and dual-task eye-tracking analysis is feasible for the detection of CI.

HIGHLIGHTS

This is the first study to report the efficiency of integrated parameters of dual-task gait and eye-tracking for cognitive impairment (CI) detection in a large cohort. We identified 46 gait and eye-tracking features associated with CI, and two were correlated to plasma phosphorylated tau 181. We constructed the model based on dual-task gait, smooth pursuit, prosaccade, and anti-saccade, achieving the best area under the curve of 0.987 for CI detection.

Naturalistic Eye Movement Tasks in Parkinson's Disease: A Systematic Review

Background

Eye tracking assessments in the laboratory have previously highlighted clear differences in eye movements between Parkinson's disease (PD) and healthy aging. However, laboratory-based eye movement tasks are artificial and limit the ecological validity of observed results. Eye movement tasks utilizing more naturalistic scenarios may provide more accurate insight into cognitive function but research in this area is limited.

Objective

This systematic review aims to ascertain what naturalistic tasks have revealed about oculomotor deficits in PD and what this information may help us understand about the underlying sensorimotor and cognitive processes.

Methods

Adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, a literature search of PsycInfo, Medline, Scopus, and Web of Science was conducted using predetermined search terms. Articles including both individuals with PD and healthy older adults completing eye tracking tasks involving naturalistic eye movements (e.g., reading, video-watching, unrestricted visual search) or naturalistic stimuli were included.

Results

After screening, 30 studies were identified as matching the inclusion criteria. Results revealed consistent findings across tasks, including longer fixation durations and smaller saccadic amplitudes in PD compared to healthy aging. However, inconsistencies in the literature and a lack of standardization in tasks limit interpretation of these results.

Conclusions

Naturalistic eye movement tasks highlight some consistent differences in eye movements between people with PD and healthy aging. However, future research should expand the current literature in this area and strive towards standardization of naturalistic tasks that can preferably be conducted remotely.

Methodologies for assessing morphosyntactic ability in people with Alzheimer's disease

BACKGROUND

The detection and description of language impairments in neurodegenerative diseases like Alzheimer's Disease (AD) play an important role in research, clinical diagnosis and intervention. Various methodological protocols have been implemented for the assessment of morphosyntactic abilities in AD; narrative discourse elicitation tasks and structured experimental tasks for production, offline and online structured experimental tasks for comprehension. Very few studies implement and compare different methodological protocols; thus, little is known about the advantages and disadvantages of each methodology.

AIMS

To discuss and compare the main behavioral methodological approaches and tasks that have been used in psycholinguistic research to assess different aspects of morphosyntactic production and comprehension in individuals with AD at the word and sentence levels.

METHODS

A narrative review was conducted through searches in the scientific databases Google Scholar, Scopus, Science Direct, MITCogNet, PubMed. Only studies written in English, that reported quantitative data and were published in peer-reviewed journals were considered with respect to their methodological protocol. Moreover, we considered studies that reported research on all stages of the disease and we included only studies that also reported results of a healthy control group. Studies that implemented standardized assessment tools were not considered in this review.

OUTCOMES & RESULTS

The main narrative discourse elicitation tasks implemented for the assessment of morphosyntactic production include interviews, picture-description and story narration, whereas the main structured experimental tasks include sentence completion, constrained sentence production, sentence repetition and naming. Morphosyntactic comprehension in AD has been assessed with the use of structured experimental tasks, both offline (sentence-picture matching, grammaticality judgment) and online (cross-modal naming,speeded sentence acceptability judgment, auditory moving window, word detection, reading). For each task we considered studies that reported results from different morphosyntactic structures and phenomena in as many different languages as possible.

CONCLUSIONS & IMPLICATIONS

Our review revealed strengths and weaknesses of these methods but also directions for future research. Narrative discourse elicitation tasks as well as structured experimental tasks have been used in a variety of languages, and have uncovered preserved morphosyntactic production but also deficits in people with AD. A combination of narrative discourse elicitation and structured production tasks for the assessment of the same morphosyntactic structure has been rarely used. Regarding comprehension, offline tasks have been implemented in various languages, whereas online tasks have been mainly used in English. Offline and online experimental paradigms have often produced contradictory results even within the same study. The discrepancy between the two paradigms has been attributed to the different working memory demands they impose to the comprehender or to the different parsing processes they tap. Strengths and shortcomings of each methodology are summarized in the paper, and comparisons between different tasks are attempted when this is possible. Thus, the paper may serve as a methodological guide for the study of morphosyntax in AD and possibly in other neurodegenerative diseases.

WHAT THIS PAPER ADDS

What is already known on this subject For the assessment of morphosyntactic abilities in AD, various methodological paradigms have been implemented: narrative discourse elicitation tasks and structured experimental tasks for production, and offline and online structured experimental tasks for comprehension. Very few studies implement and compare different methodological protocols; thus, little is known about the advantages and disadvantages of each methodology. What this paper adds to existing knowledge The paper presents an overview of methodologies that have been used to assess morphosyntactic production and comprehension of people with AD at the word and sentence levels. The paper summarizes the strengths and shortcomings of each methodology, providing both the researcher and the clinician with some directions in their endeavour of investigating language in AD. Also, the paper highlights the need for further research that will implement carefully scrutinized tasks from various experimental paradigms and will explore distinct aspects of the AD patients' morphosyntactic abilities in typologically different languages. What are the potential or actual clinical implications of this work? The paper may serve as a reference point for (psycho-)linguists who wish to study morphosyntactic abilities in AD, and for speech and language therapists who might need to apply morphosyntactic protocols to their patients in order to assess them or design appropriate therapeutic interventions for production and comprehension deficits.

Eye-tracking paradigms for the assessment of mild cognitive impairment: a systematic review

Mild cognitive impairment (MCI), representing the 'transitional zone' between normal cognition and dementia, has become a novel topic in clinical research. Although early detection is crucial, it remains logistically challenging at the same time. While traditional pen-and-paper tests require in-depth training to ensure standardized administration and accurate interpretation of findings, significant technological advancements are leading to the development of procedures for the early detection of Alzheimer's disease (AD) and facilitating the diagnostic process. Some of the diagnostic protocols, however, show significant limitations that hamper their widespread adoption. Concerns about the social and economic implications of the increasing incidence of AD underline the need for reliable, non-invasive, cost-effective, and timely cognitive scoring methodologies. For instance, modern clinical studies report significant oculomotor impairments among patients with MCI, who perform poorly in visual paired-comparison tasks by ascribing less attentional resources to novel stimuli. To accelerate the Global Action Plan on the Public Health Response to Dementia 2017-2025, this work provides an overview of research on saccadic and exploratory eye-movement deficits among older adults with MCI. The review protocol was drafted based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Electronic databases were systematically searched to identify peer-reviewed articles published between 2017 and 2022 that examined visual processing in older adults with MCI and reported gaze parameters as potential biomarkers. Moreover, following the contemporary trend for remote healthcare technologies, we reviewed studies that implemented non-commercial eye-tracking instrumentation in order to detect information processing impairments among the MCI population. Based on the gathered literature, eye-tracking-based paradigms may ameliorate the screening limitations of traditional cognitive assessments and contribute to early AD detection. However, in order to translate the findings pertaining to abnormal gaze behavior into clinical applications, it is imperative to conduct longitudinal investigations in both laboratory-based and ecologically valid settings.

Eye Movement Latency Coefficient of Variation as a Predictor of Cognitive Impairment: An Eye Tracking Study of Cognitive Impairment

Studies demonstrated impairment in the control of saccadic eye movements in Alzheimer's disease (AD) and people with mild cognitive impairment (MCI) when conducting the pro-saccade and antisaccade tasks. Research showed that changes in the pro and antisaccade latencies may be particularly sensitive to dementia and general executive functioning. These tasks show potential for diagnostic use, as they provide a rich set of potential eye tracking markers. One such marker, the coefficient of variation (CV), is so far overlooked. For biological markers to be reliable, they must be able to detect abnormalities in preclinical stages. MCI is often viewed as a predecessor to AD, with certain classifications of MCI more likely than others to progress to AD. The current study examined the potential of CV scores on pro and antisaccade tasks to distinguish participants with AD, amnestic MCI (aMCI), non-amnesiac MCI (naMCI), and older controls. The analyses revealed no significant differences in CV scores across the groups using the pro or antisaccade task. Antisaccade mean latencies were able to distinguish participants with AD and the MCI subgroups. Future research is needed on CV measures and attentional fluctuations in AD and MCI individuals to fully assess this measure's potential to robustly distinguish clinical groups with high sensitivity and specificity.

Early detection of cognitive decline in Alzheimer's disease using eye tracking

Background

Patients with Alzheimer's disease (AD) are known to exhibit visuospatial processing impairment, as reflected in eye movements from the early stages of the disease. We investigated whether the pattern of gaze exploration during visual tasks could be useful for detecting cognitive decline at the earliest stage.

Methods

Sixteen AD patients (age: 79.1 ± 7.9 years, Mini Mental State Examination [MMSE] score: 17.7 ± 5.3, mean ± standard deviation) and 16 control subjects (age: 79.4 ± 4.6, MMSE score: 26.9 ± 2.4) participated. In the visual memory task, subjects memorized presented line drawings for later recall. In the visual search tasks, they searched for a target Landolt ring of specific orientation (serial search task) or color (pop-out task) embedded among arrays of distractors. Using video-oculography, saccade parameters, patterns of gaze exploration, and pupil size change during task performance were recorded and compared between AD and control subjects.

Results

In the visual memory task, the number of informative regions of interest (ROIs) fixated was significantly reduced in AD patients compared to control subjects. In the visual search task, AD patients took a significantly longer time and more saccades to detect the target in the serial but not in pop-out search. In both tasks, there was no significant difference in the saccade frequency and amplitude between groups. On-task pupil modulation during the serial search task was decreased in AD. The number of ROIs fixated in the visual memory task and search time and saccade numbers in the serial search task differentiated both groups of subjects with high sensitivity, whereas saccade parameters of pupil size modulation were effective in confirming normal cognition from cognitive decline with high specificity.

Discussion

Reduced fixation on informative ROIs reflected impaired attentional allocation. Increased search time and saccade numbers in the visual search task indicated inefficient visual processing. Decreased on-task pupil size during visual search suggested decreased pupil modulation with cognitive load in AD patients, reflecting impaired function of the locus coeruleus. When patients perform the combination of these tasks to visualize multiple aspects of visuospatial processing, cognitive decline can be detected at an early stage with high sensitivity and specificity and its progression be evaluated.

Unsupervised mobile app-based cognitive testing in a population-based study of older adults born 1944

Background

Mobile app-based tools have the potential to yield rapid, cost-effective, and sensitive measures for detecting dementia-related cognitive impairment in clinical and research settings. At the same time, there is a substantial need to validate these tools in real-life settings. The primary aim of this study was thus to evaluate the feasibility, validity, and reliability of mobile app-based tasks for assessing cognitive function in a population-based sample of older adults.

Method

A total of 172 non-demented (Clinical Dementia Rating 0 and 0.5) older participants (aged 76-77) completed two mobile app-based memory tasks-the Mnemonic Discrimination Task for Objects and Scenes (MDT-OS) and the long-term (24 h) delayed Object-In-Room Recall Task (ORR-LDR). To determine the validity of the tasks for measuring relevant cognitive functions in this population, we assessed relationships with conventional cognitive tests. In addition, psychometric properties, including test-retest reliability, and the participants' self-rated experience with mobile app-based cognitive tasks were assessed.

Result

MDT-OS and ORR-LDR were weakly-to-moderately correlated with the Preclinical Alzheimer's Cognitive Composite (PACC5) (r = 0.3-0.44, p < .001) and with several other measures of episodic memory, processing speed, and executive function. Test-retest reliability was poor-to-moderate for one single session but improved to moderate-to-good when using the average of two sessions. We observed no significant floor or ceiling effects nor effects of education or gender on task performance. Contextual factors such as distractions and screen size did not significantly affect task performance. Most participants deemed the tasks interesting, but many rated them as highly challenging. While several participants reported distractions during tasks, most could concentrate well. However, there were difficulties in completing delayed recall tasks on time in this unsupervised and remote setting.

Conclusion

Our study proves the feasibility of mobile app-based cognitive assessments in a community sample of older adults, demonstrating its validity in relation to conventional cognitive measures and its reliability for repeated measurements over time. To further strengthen study adherence, future studies should implement additional measures to improve task completion on time.

A single-center, randomized, parallel design study to evaluate the efficacy of donepezil in improving visuospatial abilities in patients with mild cognitive impairment using eye-tracker: the COG-EYE study protocol for a phase II trial

Background

Cholinesterase inhibitors (ChEIs) decrease long-term cognitive decline in patients with Alzheimer’s disease (AD); however, there is little evidence that ChEIs affect cognitive test scores in patients with mild cognitive impairment (MCI). Conventional endpoints, such as cognitive tests or clinical rating scores, may lack the sensitivity to subtle treatment effects in participants with MCI. Therefore, there is an immediate need to refocus on direct physiological assessments to detect the effects of ChEIs in patients with MCI due to AD.

Methods

We propose a randomized controlled trial to evaluate the effect of donepezil, a ChEI, on patients with MCI due to AD. We plan to recruit 78 participants (39 in each arm) with MCI who had amyloid positron emission tomography (PET)-positive results for this open-label study. To evaluate subtle differences, we will measure eye-tracking metrics and digital pen data while participants perform the simplified Rey Complex Figure (RCFT) and clock drawing tests. The primary outcome is a change in the ratio of the number of fixations (working space/perceptual space) performed using the simplified RCFT, from baseline to 12 weeks, as assessed using eye-tracking metrics. The secondary outcomes are changes in general cognition, clinical severity, activities of daily living, and visuospatial function assessed using standard rating scores and digital pen data. The analyses of the primary and secondary outcomes will be based on the difference in changes during follow-up between the donepezil and control groups using the t-test or Mann–Whitney U test, as well as adjusting for baseline values.

Discussion

This study is designed to determine whether eye-tracking metrics can detect the effect of donepezil on visuospatial dysfunction more sensitively in patients with MCI. It is expected that multimodal data, such as eye-tracking and digital pen data, may provide helpful biomarkers for identifying subtle changes that are difficult to measure using conventional methods.

Trial registration

Clinical Research Information Service, Republic of Korea (CRIS, cris.nih.go.kr) KCT0006236. Registered on June 10, 2021.

A novel deep learning approach for diagnosing Alzheimer's disease based on eye-tracking data

Eye-tracking technology has become a powerful tool for biomedical-related applications due to its simplicity of operation and low requirements on patient language skills. This study aims to use the machine-learning models and deep-learning networks to identify key features of eye movements in Alzheimer's Disease (AD) under specific visual tasks, thereby facilitating computer-aided diagnosis of AD. Firstly, a three-dimensional (3D) visuospatial memory task is designed to provide participants with visual stimuli while their eye-movement data are recorded and used to build an eye-tracking dataset. Then, we propose a novel deep-learning-based model for identifying patients with Alzheimer's Disease (PwAD) and healthy controls (HCs) based on the collected eye-movement data. The proposed model utilizes a nested autoencoder network to extract the eye-movement features from the generated fixation heatmaps and a weight adaptive network layer for the feature fusion, which can preserve as much useful information as possible for the final binary classification. To fully verify the performance of the proposed model, we also design two types of models based on traditional machine-learning and typical deep-learning for comparison. Furthermore, we have also done ablation experiments to verify the effectiveness of each module of the proposed network. Finally, these models are evaluated by four-fold cross-validation on the built eye-tracking dataset. The proposed model shows 85% average accuracy in AD recognition, outperforming machine-learning methods and other typical deep-learning networks.

Computer-based Eye-tracking Analysis of King-Devick Test Differentiates Persons With Idiopathic Normal Pressure Hydrocephalus From Cognitively Unimpaired

BACKGROUND

Functional defects in eye movements and reduced reading speed in neurodegenerative diseases represent a potential new biomarker to support clinical diagnosis. We investigated whether computer-based eye-tracking (ET) analysis of the King-Devick (KD) test differentiates persons with idiopathic normal pressure hydrocephalus (iNPH) from cognitively unimpaired [control (CO)] and persons with Alzheimer's disease (AD).

METHODS

We recruited 68 participants (37 CO, 10 iNPH, and 21 AD) who underwent neurological examination, the Consortium to Establish a Registry for Alzheimer's Disease neuropsychological test battery (CERAD-NB), and a Clinical Dementia Rating interview. The KD reading test was performed using computer-based ET. We analyzed the total time used for the reading test, number of errors, durations of fixation and saccade, and saccade amplitudes.

RESULTS

The iNPH group significantly differed from the CO group in the KD test mean total time (CO 69.3 s, iNPH 87.3 s; P ≤0.009) and eye-tracking recording of the mean saccade amplitude (CO 3.6 degree, iNPH 3.2 degree; P ≤0.001). The AD group significantly differed from the CO group in each tested parameter. No significant differences were detected between the iNPH and AD groups.

CONCLUSION

For the first time, we demonstrated altered reading ability and saccade amplitudes in patients with iNPH.