Analysis Of Gaze Data Research Articles

Machine Learning Analysis of Gaze Data for Enhanced Precision in Diagnosing Oral Mucosal Diseases.

The diagnosis of oral mucosal diseases is a significant challenge due to their diverse differential characteristics. Risk assessment of lesions by visual examination is a complex process due to the lack of definitive guidelines. This study aimed to improve this process by creating a diagnostic algorithm using gaze data acquired during oral mucosal disease examinations. A total of 78 dentists were included in this study. Tobii Pro Nano® (Tobii Technology) was used to acquire gaze data during clinical photographic visual examinations. Advanced analysis tools such as support vector machines and heatmaps were used to visualize the gazing tendencies of a group of skilled oral surgeons, focusing on the number of gazes per region and the gazing time ratios. The preliminary findings showed the possibility of visualizing gazing tendencies and identifying areas of importance for diagnosis. The classification of intraoral photographs based on gross features revealed the existence of an optimal examination method for each category and diagnostically significant areas. This novel approach to analyzing gaze data has the potential to refine diagnostic techniques and increase both accuracy and efficiency.

Fast mapping in word-learning: A case study on the humanoid social robots' impacts on Children's performance

The evolution of robotic technology has heightened the demands for robots capable of social interaction with humans, also known as social robots. The encouraging prospects of deploying this advanced assistive tool in entertainment, healthcare, and education have, in turn, promoted substantial growth in studies conducted on the human-robot interaction (HRI) topic. One hot HRI subject is the examination of social robots' efficacy within various educational scenarios, e.g., word learning. In the context of word learning, fast mapping (FM) is a well-established technique through which novel words are taught by contrasting them with previously known words. The concept of this approach originates from how infants comprehend new vocabularies' referents. In this study, we investigated the potential benefits of utilizing social robots during educational interventions, based on the FM strategy, by means of two measures, alternative forced-choice (AFC) test and gaze data analysis. The first metric explores participants' retention after a 10-min and a one-week delay, while the second is concerned with scrutinizing their gaze focus during training sessions. Our findings indicate that the utility of social robots during FM interventions could enhance children's engagement, attention, and retention of novel words.

Reflexive Saccades Used for Objective and Automated Measurements of Contrast Sensitivity in Selected Areas of Visual Field

PurposeThis study proposes a novel approach for objective and automated peripheral contrast sensitivity (CS) testing using reflexive saccades. Here the CS was examined in various areas the of visual field (VF) using a live analysis of gaze data. For validation of the new test, we examined CS with an established procedure of identifying the orientation of a contrast stimulus.MethodsTo perform and validate the saccade-based testing, two separate measurement events were performed. In the first, participants were asked to execute a saccade toward a newly-appeared stimulus in their VF. After the saccade execution or stimulus expiry, reporting the target orientation was required in a four-alternatives forced choice (4AFC). Therefore the first measurement yields two outcomes (objective and subjective). In the second measurement, only the identification of the stimulus orientation was requested, while fixating a central mark. Stimulus contrast was controlled by an adaptive psychometric procedure in both measurements.ResultsThe study found strong correlations (all r ≥ 0.79) of CS values for all three possible testing methods (saccade-based responding in saccadic measurements, keyboard-based responding in saccadic measurements, keyboard-based responding in non-saccadic measurements), showing the feasibility of employment of reflexive saccades in such testing. Second, this study shows a significant influence of eccentricity and direction of the stimulus on the CS function.ConclusionsCS measured with reflexive saccades is comparable to other testing methods over several areas of the participant's VF. Hence, we propose it as a novel and objective testing procedure for CS measurements.Translational RelevanceAssessment of CS using reflexive saccades extends the portfolio of suggested eye movement-based tests, allowing objective examination across the VF, which might be helpful especially in the early detection of various eye diseases.

Data-driven analysis of gaze patterns in face perception: Methodological and clinical contributions

Gaze patterns during face perception have been shown to relate to psychiatric symptoms. Standard analysis of gaze behavior includes calculating fixations within arbitrarily predetermined areas of interest. In contrast to this approach, we present an objective, data-driven method for the analysis of gaze patterns and their relation to diagnostic test scores. This method was applied to data acquired in an adult sample (N = 111) of psychiatry outpatients while they freely looked at images of human faces. Dimensional symptom scores of autism, attention deficit, and depression were collected. A linear regression model based on Principal Component Analysis coefficients computed for each participant was used to model symptom scores. We found that specific components of gaze patterns predicted autistic traits as well as depression symptoms. Gaze patterns shifted away from the eyes with increasing autism traits, a well-known effect. Additionally, the model revealed a lateralization component, with a reduction of the left visual field bias increasing with both autistic traits and depression symptoms independently. Taken together, our model provides a data-driven alternative for gaze data analysis, which can be applied to dimensionally-, rather than categorically-defined clinical subgroups within a variety of contexts. Methodological and clinical contribution of this approach are discussed.

Value of Eye-Tracking Data for Classification of Information Processing-Intensive Handling Tasks: Quasi-Experimental Study on Cognition and User Interface Design.

BackgroundIn order to give a wide range of people the opportunity to ensure and support home care, one approach is to develop medical devices that are as user-friendly as possible. This allows nonexperts to use medical devices that were originally too complicated to use. For a user-centric development of such medical devices, it is essential to understand which user interface design best supports patients, caregivers, and health care professionals.ObjectiveUsing the benefits of mobile eye tracking, this work aims to gain a deeper understanding of the challenges of user cognition. As a consequence, its goal is to identify the obstacles to the usability of the features of two different designs of a single medical device user interface. The medical device is a patient assistance device for home use in peritoneal dialysis therapy.MethodsA total of 16 participants, with a subset of seniors (8/16, mean age 73.7 years) and young adults (8/16, mean age 25.0 years), were recruited and participated in this study. The handling cycle consisted of seven main tasks. Data analysis started with the analysis of task effectiveness for searching for error-related tasks. Subsequently, the in-depth gaze data analysis focused on these identified critical tasks. In order to understand the challenges of user cognition in critical tasks, gaze data were analyzed with respect to individual user interface features of the medical device system. Therefore, it focused on the two dimensions of dwell time and fixation duration of the gaze.ResultsIn total, 97% of the handling steps for design 1 and 96% for design 2 were performed correctly, with the main challenges being task 1 insert, task 2 connect, and task 6 disconnect for both designs. In order to understand the two analyzed dimensions of the physiological measurements simultaneously, the authors propose a new graphical representation. It distinguishes four different patterns to compare the eye movements associated with the two designs. The patterns identified for the critical tasks are consistent with the results of the task performance.ConclusionsThis study showed that mobile eye tracking provides insights into information processing in intensive handling tasks related to individual user interface features. The evaluation of each feature of the user interface promises an optimal design by combining the best found features. In this way, manufacturers are able to develop products that can be used by untrained people without prior knowledge. This would allow home care to be provided not only by highly qualified nurses and caregivers, but also by patients themselves, partners, children, or neighbors.

The Visual Search Strategies Underpinning Effective Observational Analysis in the Coaching of Climbing Movement.

Despite the importance of effective observational analysis in coaching the technical aspects of climbing performance, limited research informs this aspect of climbing coach education. Thus, the purpose of the present research was to explore the feasibility and the utility of a novel methodology, combining eye tracking technology and cued retrospective think-aloud (RTA), to capture the cognitive–perceptual mechanisms that underpin the visual search behaviors of climbing coaches. An analysis of gaze data revealed that expert climbing coaches demonstrate fewer fixations of greater duration and fixate on distinctly different areas of the visual display than their novice counterparts. Cued RTA further demonstrated differences in the cognitive–perceptual mechanisms underpinning these visual search strategies, with expert coaches being more cognizant of their visual search strategy. To expand, the gaze behavior of expert climbing coaches was underpinned by hierarchical and complex knowledge structures relating to the principles of climbing movement. This enabled the expert coaches to actively focus on the most relevant aspects of a climber’s performance for analysis. The findings demonstrate the utility of combining eye tracking and cued RTA interviewing as a new, efficient methodology of capturing the cognitive–perceptual processes of climbing coaches to inform coaching education/strategies.

MAD saccade: statistically robust saccade threshold estimation via the median absolute deviation.

Saccade detection is a critical step in the analysis of gaze data. A common method for saccade detection is to use a simple threshold for velocity or acceleration values, which can be estimated from the data using the mean and standard deviation. However, this method has the downside of being influenced by the very signal it is trying to detect, the outlying velocities or accelerations that occur during saccades. We propose instead to use the median absolute deviation (MAD), a robust estimator of dispersion that is not influenced by outliers. We modify an algorithm proposed by Nyström and colleagues, and quantify saccade detection performance in both simulated and human data. Our modified algorithm shows a significant and marked improvement in saccade detection - showing both more true positives and less false negatives – especially under higher noise levels. We conclude that robust estimators can be widely adopted in other common, automatic gaze classification algorithms due to their ease of implementation.

MLGaze: Machine Learning-Based Analysis of Gaze Error Patterns in Consumer Eye Tracking Systems.

Analyzing the gaze accuracy characteristics of an eye tracker is a critical task as its gaze data is frequently affected by non-ideal operating conditions in various consumer eye tracking applications. In previous research on pattern analysis of gaze data, efforts were made to model human visual behaviors and cognitive processes. What remains relatively unexplored are questions related to identifying gaze error sources as well as quantifying and modeling their impacts on the data quality of eye trackers. In this study, gaze error patterns produced by a commercial eye tracking device were studied with the help of machine learning algorithms, such as classifiers and regression models. Gaze data were collected from a group of participants under multiple conditions that commonly affect eye trackers operating on desktop and handheld platforms. These conditions (referred here as error sources) include user distance, head pose, and eye-tracker pose variations, and the collected gaze data were used to train the classifier and regression models. It was seen that while the impact of the different error sources on gaze data characteristics were nearly impossible to distinguish by visual inspection or from data statistics, machine learning models were successful in identifying the impact of the different error sources and predicting the variability in gaze error levels due to these conditions. The objective of this study was to investigate the efficacy of machine learning methods towards the detection and prediction of gaze error patterns, which would enable an in-depth understanding of the data quality and reliability of eye trackers under unconstrained operating conditions. Coding resources for all the machine learning methods adopted in this study were included in an open repository named MLGaze to allow researchers to replicate the principles presented here using data from their own eye trackers.

GazeVisual: A Practical Software Tool and Web Application for Performance Evaluation of Eye Tracking Systems

The concept and functionalities of a software tool developed for in depth performance evaluation of eye gaze estimation systems is presented. The software, GazeVisual has capabilities for quantitative, statistical, and visual analysis of eye gaze data as well as generation of static and dynamic visual stimuli for sample gaze data collection. This is a first of its kind cross-platform tool for gaze data analysis and evaluation. This software is made freely available to the eye gaze research and development community to provide a common framework for estimating the quality and reliability of data from eye tracking systems, especially those implemented in consumer electronics (CE) applications. The feasibility of using this software is tested through case studies which show that the software can handle eye gaze datasets obtained from several different consumer grade eye trackers. GazeVisual operates consistently, irrespective of the platform, algorithm or hardware of the eye trackers. In addition, the GazeVisual software capabilities are also made accessible via a Web-based application enabling performance evaluation of eye tracker data over a cloud-based platform.

Fixation data analysis for complex high-resolution satellite images*

This paper reports an empirical study which inspects the human segmentation process for high-resolution satellite (HRS) images in the absence of target objects. The lack of standard guidelines for eye-tracking data analysis in the above-mentioned case motivates this study. Since a careful selection of area of interest (AOI) is necessary for meaningful interpretation of gaze data, this study represents a way to draw AOIs which metrics correlate with the human verbal response. This study formulates a hypothesis for AOI drawing and updates it until the metrics of the hypothesis based AOIs matches with the verbal response. The outputs of this study suggest the grouping of objects during AOI-based analysis of gaze data and provides some rules for the same. The AOIs drawing for hypothesis validation also highlights the dominance of perceptual grouping over the functional grouping of objects. Developing an autonomous system for segmentation and reference data creation are some possible future scopes of this study.

Automatic Query Refining Based on Eye-Tracking Feedback

This paper presents a new method named AQueReBET, which automatically refines a query set by an information seeker searching on the web. A revelation of the intention of an information seeker who is running a search can bring a significant improvement to the search process, and to browsing as well. It is practically impossible to acquire such intention by the explicit indication (feedback) due to the fact that web browsing takes place in real time. Therefore the intention must be determined in some other way. We hypothesize that it can be approximated by means of the implicit feedback preferably in the form of data from an eye tracker and mouse. We propose a method which automatically refines a seeker’s search query and thus we can offer documents with higher relevance, decrease the number of query reformulations and increase the seeker’s satisfaction. The query refinement is based on an analysis of gaze data from an eye tracker and also on groupization. In the proposed method, we calculate word-level importance based on term frequency, term uniqueness (tf-idf) and total fixation duration within the subdocument (word's snippet in search results).

Screen recording segmentation to scenes for eye-tracking analysis

In usability studies involving eye-tracking, quantitative analysis of gaze data requires the information about so called scene occurrences. Scene ocurrences are time segments during which the application user interface remains more-less static, so gaze events (e.g., fixations) can be mapped to the particular areas of interest (user interface elements). The scene occurrences typically start and end by user interface changes such as page-to-page transitions, menu expansions, overlay propmts, etc. Normally, one would record such changes programmatically through application logging, yet in many studies, this is not possible. For example, in an early-prototype mobile-app testing, only a camera recording of a smart device screen is often available as evidence. In such cases, analysts must manually annotate the recordings. To reduce the need for manual annotation of scene occurrences, we present an image processing method for segmenting user interface video recordings. The method exploits specific properties of user interface recordings, which greatly differ from real world video shots (for which many segmentation methods exist). The core of our method lies in the use of SSIM and SIFT similarity metrics used on video frames (with several pre-processing and filtering procedures). The main advantage of our method is, that it requires no training data apart from single screenshot example for each scene (to which the recording frames are compared). The method is also able to work with user finger overlays, which are always present in mobile device recordings. We evaluate the accuracy of our method over recordings from several real-life studies and compare it with other image similarity techniques.

A low-cost platform for eye-tracking research: Using Pupil© in behavior analysis.

Tracking eye movements is being increasingly recognized as a valuable source of information about stimulus control. So far, however, eye-tracking research has suffered from accessibility issues, with expensive hardware and closed-source software. In this article we review Pupil©, an eye-tracking platform developed by Pupil Labs and that combines open-source software with low-cost hardware components. We offer concrete recommendations about Pupil use in stimulus-control research and we show how the software can be extended to automatize the analysis of gaze data. Finally, we present the results of a study of visual discrimination and conditioned reinforcement conducted with Pupil, establishing the usefulness of this platform as a research tool in behavior analysis.

Eye tracking in catheter-based cardiovascular interventions: early results.

.Visual x-ray image processing (XRIP) represents a fundamental component of catheter-based cardiovascular interventions (CBCVIs). To date, no data are available to define XRIP in this setting. To characterize CBCVI XRIP, we developed a computer-based method allowing continuous temporal–spatial analysis of data recorded by a head-mounted eye-tracking device. Quantitative analysis of gaze duration of an expert operator (EO) revealed that the average time in minutes spent viewing the images on the display screen was and of the total recorded time in coronary angiography (CA) and in CA followed by CBCVI, respectively. Qualitative analysis of gaze data of the EO revealed consistent focus on the center point of the screen. Only if suspicious findings were detected did gaze move toward the target. In contrast, a novice operator (NO) observing a subset of cases viewed coronary artery segments separately and sequentially. The developed methodology allows continuous registration and analysis of gaze data for analysis of XRIP strategies of EOs in live-cases scenarios and may assist in the transfer of experts’ reading skills to novices.

GazeDx: Interactive Visual Analytics Framework for Comparative Gaze Analysis with Volumetric Medical Images.

We present an interactive visual analytics framework, GazeDx (abbr. of GazeDiagnosis), for the comparative analysis of gaze data from multiple readers examining volumetric images while integrating important contextual information with the gaze data. Gaze pattern comparison is essential to understanding how radiologists examine medical images, and to identifying factors influencing the examination. Most prior work depended upon comparisons with manually juxtaposed static images of gaze tracking results. Comparative gaze analysis with volumetric images is more challenging due to the additional cognitive load on 3D perception. A recent study proposed a visualization design based on direct volume rendering (DVR) for visualizing gaze patterns in volumetric images; however, effective and comprehensive gaze pattern comparison is still challenging due to a lack of interactive visualization tools for comparative gaze analysis. We take the challenge with GazeDx while integrating crucial contextual information such as pupil size and windowing into the analysis process for more in-depth and ecologically valid findings. Among the interactive visualization components in GazeDx, a context-embedded interactive scatterplot is especially designed to help users examine abstract gaze data in diverse contexts by embedding medical imaging representations well known to radiologists in it. We present the results from two case studies with two experienced radiologists, where they compared the gaze patterns of 14 radiologists reading two patients' volumetric CT images.

ICOMPONENT: SOFTWARE WITH FLEXIBLE ARCHITECTURE FOR DEVELOPING PLUG-IN MODULES FOR EYE TRACKERS

iComponent is a software tool we have developed to facilitate basic research on eye movements as well as applications of eye gaze for computer input. Despite the variety of eye trackers and gaze-data analysis tools avail- able today, there is still a gap between what researchers need and what products are available on the market to suit their needs. Also, researchers are confronted with many difficulties due to complexity of the technology of eye tracking de- vices. We approached the need to standardize the input process by proposing a generic data format and making the software architecture flexible. This is achieved by inserting an intermediate software layer (API converter) between the core of iComponent and the API of the device attached.

Three-dimensional information in face recognition: An eye-tracking study

One unresolved question about face perception is: what is the role of three-dimensional information in face recognition? In this study, recognition performance was compared across changes in viewpoint in different depth conditions: a 2D condition without stereo information and a 3D condition where stereo information was present (by viewing the same face images as anaglyphs through 3D glasses). Subjects' eye movements were recorded during both 3D and 2D sessions. The findings revealed that participants were more accurate in the 3D condition. Moreover, individual differences in interpupillary distance predicted recognition performance in the 3D but not in the 2D condition. A "region of interest" analysis of gaze data showed that rich volumetric properties provided by certain facial features (e.g., the nose and the cheeks) were attended more in the 3D condition compared to the 2D condition. Taken together, these findings support the conclusion that face recognition across viewpoint transformation is facilitated by the addition of stereoscopic depth cues.

Fixation identification: The optimum threshold for a dispersion algorithm

It is hypothesized that the number, position, size, and duration of fixations are functions of the metric used for dispersion in a dispersion-based fixation detection algorithm, as well as of the threshold value. The sensitivity of the I-DT algorithm for the various independent variables was determined through the analysis of gaze data from chess players during a memory recall experiment. A procedure was followed in which scan paths were generated at distinct intervals in a range of threshold values for each of five different metrics of dispersion. The percentage of points of regard (PORs) used, the number of fixations returned, the spatial dispersion of PORs within fixations, and the difference between the scan paths were used as indicators to determine an optimum threshold value. It was found that a fixation radius of 1 degrees provides a threshold that will ensure replicable results in terms of the number and position of fixations while utilizing about 90% of the gaze data captured.