Eye Movement Classification Research Articles

I04 Palm-size oculomotor laboratory for field testing of saccadic system potentially applicable for monitoring of HD development

Saccadic eye movement measurement is reported as potential sensitive biomarker of Huntington's disease (HD), in both pre-manifest and manifest stages. While definitely less specific than imagining techniques, it possess potential for quick, low cost and non invasive quantitative assessment of brain functional status. It is also less prone to neuronal compensation mechanism which continues to conceal the evidence of functional motor impairment, especially when voluntary component is involved in saccadic test paradigm (antisaccade-peripherial-conflict). According to recent findings, the changes of saccadic metrics accompany HD progression and they can be considered as a means of tracking the current patient condition and outcome of new therapeutic interventions. Authors present portable, integrated diagnostic environment, Saccadometer Clinical. The system is designed as minimal-setup for evaluation of saccadic metrics, allowing to establish subjects functional deterioration profile based on the repetitive measurements (status samples) accumulated over the time. The key component is small factor, battery powered, head-mounted intelligent sensor, capable of generating laser dot stimuli for both reflex and volitional decision based saccadic tests, and performing automated eye movement acquisition and classification. Due to portability, ease of use, and build-in telemetry link for wireless communication, the measurement can be carried by patient her/himself in home environment, providing facility for continuous, remote monitoring of patient's functional status. In order to broaden the spectrum of measured parameters, current development efforts aim to include the smooth pursuit and the vestibule-ocular (postural sway) tests within the same miniaturised system, serving as complementary biomarkers of HD.

Automated classification and scoring of smooth pursuit eye movements in the presence of fixations and saccades

Ternary eye movement classification, which separates fixations, saccades, and smooth pursuit from the raw eye positional data, is extremely challenging. This article develops new and modifies existing eye-tracking algorithms for the purpose of conducting meaningful ternary classification. To this end, a set of qualitative and quantitative behavior scores is introduced to facilitate the assessment of classification performance and to provide means for automated threshold selection. Experimental evaluation of the proposed methods is conducted using eye movement records obtained from 11 subjects at 1000 Hz in response to a step-ramp stimulus eliciting fixations, saccades, and smooth pursuits. Results indicate that a simple hybrid method that incorporates velocity and dispersion thresholding allows producing robust classification performance. It is concluded that behavior scores are able to aid automated threshold selection for the algorithms capable of successful classification.

Evaluating Feature Extraction Methods of Electrooculography (EOG) Signal for Human-Computer Interface

Electrooculography (EOG) signal is a widely and successfully used to detect activities of human eye. Use of the EOG signals as a control signal for human-computer interface (HCI) plays a central role in the understanding, characterization and classification of eye movements which can be applied to a wide variety of applications consisting virtual mouse and keyboard control, electric power wheelchairs and industrial assistive robots. The advantages of the EOG-based interface over other conventional interfaces have been presented in the last two decades; however, due to a lot of information in EOG signals, the extraction of useful features should be done before the classification task. In this study, fourteen useful features extracted from two directional EOG signals: vertical (V) and horizontal (H) signals have been presented and evaluated. There are the maximum peak and valley amplitude values (PAV and VAV), the maximum peak and valley position values (PAP and VAP), the area under curve value (AUC), the number of threshold crossing value (TCV), and EOG variance (VAR), which are derived from both V and H signals. In the experiments, EOG signals obtained from three healthy subjects with eight directional eye movements were employed: up, down, right, left, up-right, up-left, down-right and down-left. The mean feature values and their standard deviations have been reported. Most features show the difference between the mean feature values. Using the analysis-of-variation test, the differences in mean features between the movements are statistically significant for ten features (p < 0.0001), particularly for the VAV, VAP, AUC, TCV and VAR of V signal, and the PAV, VAV, AUC, TCV and VAR of H signal. The combination of these features may be useful for the classification of EOG signals in both class separability and robustness point of views. Using multiple features with sufficient classifiers or threshold techniques is recommended to be evaluated in further analysis. These features can be useful for various advanced HCI applications in future researches, notably eye-exercise and eye-writing recognitions.

Standardization of Automated Analyses of Oculomotor Fixation and Saccadic Behaviors

In an effort towards standardization, this paper evaluates the performance of five eye movement classification algorithms in terms of their assessment of oculomotor fixation and saccadic behavior. The results indicate that performance of these five commonly used algorithms vary dramatically even in the case of a simple stimulus evoked task using a single, common threshold value. The important contributions of this paper are: 1) evaluation and comparison of performance of five algorithms to classify specific oculomotor behavior 2) introduction and comparison of new standardized scores to provide more reliable classification performance 3) logic for a reasonable threshold value selection for any eye movement classification algorithm based on the standardized scores and 4) logic for establishing a criterion-based baseline for performance comparison between any eye movement classification algorithms. Proposed techniques enable efficient and objective clinical applications providing means to assure meaningful automated eye movement classification.

On‐Line Classification and Prediction of Eye Movements by Multiple‐Model Kalman Filtering

An extensible multiple-model Kalman filter framework for eye tracking and video-oculography (VOG) applications is proposed. The Kalman filter predicts future states of a system on the basis of a mathematical model and previous measurements. The predicted values are then compared against the current measurements. In a correcting step, the predicted state is enhanced by the measurements. In this work, the Kalman filter is used for smoothing the VOG data, for on-line classification of eye movements, as well as for predictive real-time control of a gaze-driven head-mounted camera (EyeSeeCam). With multiple models running in parallel, it was possible to distinguish between fixations, slow-phase eye movements, and saccades. Under the assumption that each class of eye movement follows a distinct model, one can decide which types of eye movement occurred by evaluating the probability for each model.

Classification of Non-Nystagmic Spontaneous Pathological Eye Movements

We attempted to classify the well-known spontaneous pathological eye movement (SPEM), to provide a unified nomenclature, to describe its pathophysiology, and to determine its significance in determining a differential diagnosis. Nomenclature was according to the first reporter. These phenomena are easy to confuse with artifact when recording ENGs. A careful analysis which considers history and general neurological findings can identify many of these as true clinical findings which may be as important as nystagmus in the establishing of a clinical diagnosis. Each SPEM, once confirmed, plays a decisive role in the localization of lesions within the central nervous system.

Centering of eyes; a patterned eye movement.

IN A STUDY of patterns of eye movements obtained on electric stimulation of brain stem, cerebellum, and cerebrum in monkey, there was one type of binocular action which has seldom been mentioned in conventional classification of eye movements. This was a centering of eyes to midposition regardless of location of globes prior to stimulation. Magoun, in his earlier investigations of brain stem, recorded such eye movements in his protocols, but he did not attach particular significance to them. In our preliminary note, made in 1948, we called this ocular movement the midposition phenomenon. 1 The same phenomenon had previously been observed by others who investigated cerebral function. In 1888 Beevor and Horsley 2 and in 1890 Mott and Schaefer 3 noted this centering in their observations on stimulation of frontal lobe. Brown 4 interpreted this midpositioning as the orientation of