Gaze Metrics Research Articles

Prehospital ultrasound constitutes a potential distraction from the observation of critically ill patients: a prospective simulation study

BackgroundPrehospital point-of-care ultrasound allows an unstable patient to be rapidly and accurately assessed. However, we are concerned that an excessive focus on the ultrasound device, in an already demanding emergency medical service environment, may distract from patient care, potentially leading to reduced situational awareness and the neglect of other crucial instruments, such as the patient monitor. Thus, in this study, we examined the influence of prehospital ultrasound on situational awareness, by studying the degree to which physicians were distracted from the patient monitor.MethodsWe observed HEMS physicians in a simulated setting and analysed their gaze behaviour using an eye tracker placed on three areas of interests: the ultrasound device, the patient and the patient monitor. In the course of the experiment, the simulated patient desaturated, which was presented on the patient monitor. The primary outcome was the fraction of gaze distribution across the three areas of interest, while secondary outcomes were different gaze metrics (dwell time, revisits, average duration of visual intake and entry time) on the patient monitor. We then compared the participants who noticed the patient’s deterioration with those who did not.ResultsIn 75% of cases, the severely decreased oxygen saturation went unnoticed during the test. Moreover, the gaze distribution of the two groups differed, with the group that recognised the deterioration focusing longer on the patient monitor (7.8% (95% CI 5–10.7) vs 0.1% (95% CI 0–0.3), p: 0.124).ConclusionsThe task of performing an ultrasound examination appears to overwhelm some participants and distract them from other aspects of the scenario. Efforts to mitigate distractions and optimise the use of prehospital ultrasound, such as education, a focus on human factors aspects and standardisation, are crucial for maximising the potential benefits of prehospital ultrasound.

Flashlight model: Integrating attention distribution and attention resources for pilots’ visual behaviour analysis and performance prediction

Pilot performance is almost the last line of defense in aircraft safety. Recent years have seen a surge in research aimed at utilizing eye-tracking technology to predict pilot performance, enhancing aviation safety margins. A decline in pilot performance is often attributed to either misdirected attention towards irrelevant tasks or inefficient information processing owing to limited attention resources. Previous research has shown that eye-tracking data can effectively capture these issues and provide accurate performance predictions. Nevertheless, the existing studies either focus on attention distribution or attention resources separately, neglecting the complex interactions between them. To address this gap, our study proposes a synthesized Flashlight model-based eye-tracking analysis for pilot performance prediction, integrating the two perspectives. Accordingly, the combined AOI-gaze metrics are proposed to offer a more nuanced analysis of information processing across specific Areas of Interest (AOIs), thereby enhancing the analysis of gaze metrics. We examined the efficacy of the combined AOI-gaze metrics in the Gradient-boosted decision trees(GBDT) model for pilot performance prediction and compared them with other widely used eye-tracking metrics in a simulated flight experiment case study. Moreover, we employed the SHapley Additive exPlanations (SHAP) method to identify the most influential eye-tracking measurements for pilots’ performance prediction. The result demonstrated that the selected eye-tracking measurements obtained the highest accuracy in performance prediction.

Eye tracking during a simulated start of shift safety check: An observational analysis of gaze behavior of critical care nurses.

The handover and associated shift start checks by nurses of critical care patients are complex and prone to errors. However, which aspects lead to errors remains unknown. Fewer errors might occur in a structured approach. We hypothesized that specific gaze behavior during handover and shift start safety check correlates with error recognition. In our observational eye tracking study, we analyzed gaze behavior of critical care nurses during handover and shift start safety check in a simulation room with built-in errors. Four areas of interest (AOI) were pre-defined (patient, respirator, prescriptions, monitor). The primary outcome were different gaze metrics (time to first fixation, revisits, first visual intake duration, average visual intake duration, dwell time) on AOIs. Parameters were analyzed by taking all errors in account, and by dividing them into minor and critical. Forty-three participants were included. All participants committed at least a minor error (n = 43, 100%), at least one critical error occurred in 29 participants (67%). Taking all errors into account, longer time to first fixation and more revisits were associated with an increased risk of missing errors (Time to First Fixation: OR 1.099 (95% CI 1.023-1.191, p = 0.0002), Revisits: OR 1.080 (95% CI 1.025-1.143, p = 0.0055)). Error detection during shift start safety check was associated with distinct gaze behavior. Nurses who recognized more errors had a shorter time to first fixation and less revisits. These gaze characteristics might correspond to a more structured approach. Further research is necessary, for example by implementing a checklist, to reduce errors in the future and improve patient safety.

With sensitive eyes: ECEC teachers’ visual gaze and related reflections on pedagogical actions in toddler groups using eye-tracking glasses

This study explored early childhood education and care (ECEC) teachers’ visual gaze and related reflections on pedagogical actions during pedagogical activities in groups of children under three years of age in Finland. The data were collected from play and teacher-guided activities using mobile eye-tracking glasses, the retrospective thinking aloud (RTA) method, and semi-structured interviews. The results showed that even though the teachers were surprised about some aspects of the visual gaze metrics, they reflected on and gave reasons for their visual gazes on children. When observing gaze data from play, teachers explained the high amount of gaze by citing children’s particular needs. When observing gaze data from guided activities, teachers reflected on children’s unpredictable behavior and noted that the children’s need for support in concentration was linked to more gazes by the teacher. The findings showed that both during play and guided activities, children seeking a gaze and the position of children in the classroom influenced the number of teachers’ gazes. In the teachers’ explanations of their visual gaze and related pedagogical actions, five categories were identified: protection; physical and emotional availability, teaching and learning; facilitation; and initiatives. This explorative study showed that teachers utilize their knowledge concerning children’s individuality, development, and learning when they explain their decisions concerning their visual gaze and pedagogical activities with toddlers. The use of mobile eye-tracking technology is relatively new; therefore, its applications to ECEC are pioneering for the development of the field in relation to the practices and research of toddlers’ groups and groups with older children in ECEC

Audio augmented reality using sonification to enhance visual art experiences: Lessons learned

Augmenting visual art in art galleries can be an effective Audio Augmented Reality (AAR) application for indoor exploration. In the current study, eight paintings from four genres were augmented with audio through their sonification. Basic Audio was generated using a sonification algorithm by identifying the major colors of the paintings, and Enhanced Audio was generated by a musician enhancing the Basic Audio; these were presented with the paintings to compare against No Audio. Twenty-six participants viewed each painting in all three conditions; eye gaze metrics, and qualitative data were collected. Results showed that Enhanced Audio led to significantly greater engagement and positive sentiments, compared to Basic Audio. Thematic analysis showed semantic and syntactic relationships of the audio with the paintings, and a tendency to guide users’ gaze over time. Findings from this study can guide future AAR developments to improve auditory display designs to enhance visual experiences.

Investigating the role of flight phase and task difficulty on low-time pilot performance, gaze dynamics and subjective situation awareness during simulated flight.

Gaze behaviour has been used as a proxy for information processing capabilities that underlie complex skill performance in real-world domains such as aviation. These processes are highly influenced by task requirements, expertise and can provide insight into situation awareness (SA). Little research has been done to examine the extent to which gaze behaviour, task performance and SA are impacted by various task manipulations within the confines of early-stage skill development. Accordingly, the current study aimed to understand the impact of task difficulty on landing performance, gaze behaviour and SA across different phases of flight. Twenty-four low-time (<300 hours) pilots completed simulated landing scenarios under visual flight rules conditions. Traditional gaze metrics, entropybased metrics, and blink rate provided meaningful insight about the extent to which information processing is modulated by flight phase and task difficulty. The results also suggested that gaze behavior changes compensated for increased task demands and minimized the impact on task performance. Dynamic gaze analyses were shown to be a robust measure of task difficulty and pilot flight hours. Recommendations for the effective implementation of gaze behaviour metrics and their utility in examining information processing changes are discussed.

Pedestrians’ responses to scalable automated vehicles with different external human-machine interfaces: Evidence from a video-based eye-tracking experiment

To enhance the efficiency and safety of interactions with pedestrians, numerous external Human-Machine Interfaces (eHMIs) concepts for automated vehicles (AVs) have been proposed and evaluated, predominately based on singular pedestrian-AV interaction scenarios. This leaves a gap in comprehending the efficiency and robustness of eHMIs during interactions with scalable AVs. To bridge the gap, this study pioneers an exploration of pedestrians' road-crossing decisions, perceived clarity, and gaze behaviour during synchronous interactions with multiple AVs equipped with different eHMIs. An eye-tracking experiment was conducted, involving 48 video stimuli depicting two AVs (with yielding patterns and eHMIs manipulated) approaching a predetermined road-crossing location from the same side of a two-way, four-lane, un-signalized road. Each yielding AV will present one of four eHMI manipulations (i.e., light band, smiling expression, pedestrian symbol, and no eHMI as baseline), resulting in 16 unique eHMI combinations and video stimuli. Eighty-seven participants were recruited and tasked to respond to approaching AVs, during which their road-crossing decision time, perceived clarity, and gaze metrics were recorded. Findings revealed that different eHMIs across AVs were associated with prolonged road-crossing decision times, diminished perceived clarity, and intensified visual attention as evidenced by increased fixation counts and durations. Regarding AVs equipped with the same eHMIs, the pedestrian symbol-based eHMIs resulted in heightened clarity perception, reduced crossing decision times and visual engagement, in contrast to eHMIs presented in abstract light bands or smiling expressions. Furthermore, for eHMI-different AVs, those featuring only one eHMI within AVL-AVR pairs were perceived as less clear, eliciting heightened visual cognitive load and decision times. Integration of a pedestrian symbol-based eHMI in AVL-AVR configurations garnered superior clarity ratings and facilitated time savings. These findings of this pioneering study provide insights for eHMI design for future AVs in complex traffic scenarios with scalable road users.

Performance of the Action Observation-Based Brain-Computer Interface in Stroke Patients and Gaze Metrics Analysis.

Brain-computer interfaces (BCIs) are anticipated to improve the efficacy of rehabilitation for people with motor disabilities. However, applying BCI in clinical practice is still a challenge due to the great diversity of patients. In the current study, a novel action observation (AO) based BCI was proposed and tested on stroke patients. Ten non-hemineglect patients and ten hemineglect patients were recruited. Four AO stimuli were designed, each presenting a decomposed action to complete the reach-and-grasp task. EEG data and eye movement data were collected. Eye movement data was utilized to analyze the reasons for individual differences in BCI performance. Task discriminative component analysis was utilized to perform online target detection. The results showed that the designed AO-based BCI could simultaneously induce steady state motion visual evoked potential (SSMVEP) from the occipital region and sensory motor rhythm from the sensorimotor region in stroke patients. The average online detection accuracy among the four AO stimuli reached 67% within 3 s in the non-hemineglect group, while the accuracy only reached 35% in the hemineglect group. Gaze metrics showed that the average total duration of fixations during the stimulus phase in the hemineglect group was only 1.31 s ± 0.532 s which was significantly lower than that in the non-hemineglect group. The results indicated that hemineglect patients have difficulty gazing at the AO stimulus, resulting in poor detection performance and weak desynchronization in the sensorimotor region. Furthermore, the degree of neglect is inversely proportional to the target detection accuracy in hemineglect stroke patients. In addition, the gaze metrics associated with cognitive load were significantly correlated with the accuracy in non-hemineglect patients. It indicated the cognitive load may affect the AO-based BCI. The current study will expedite the clinical application of AO-based BCI.

Extracting decision-making features from the unstructured eye movements of clinicians on glaucoma OCT reports and developing AI models to classify expertise.

This study aimed to investigate the eye movement patterns of ophthalmologists with varying expertise levels during the assessment of optical coherence tomography (OCT) reports for glaucoma detection. Objectives included evaluating eye gaze metrics and patterns as a function of ophthalmic education, deriving novel features from eye-tracking, and developing binary classification models for disease detection and expertise differentiation. Thirteen ophthalmology residents, fellows, and clinicians specializing in glaucoma participated in the study. Junior residents had less than 1 year of experience, while senior residents had 2-3 years of experience. The expert group consisted of fellows and faculty with over 3 to 30+ years of experience. Each participant was presented with a set of 20 Topcon OCT reports (10 healthy and 10 glaucomatous) and was asked to determine the presence or absence of glaucoma and rate their confidence of diagnosis. The eye movements of each participant were recorded as they diagnosed the reports using a Pupil Labs Core eye tracker. Expert ophthalmologists exhibited more refined and focused eye fixations, particularly on specific regions of the OCT reports, such as the retinal nerve fiber layer (RNFL) probability map and circumpapillary RNFL b-scan. The binary classification models developed using the derived features demonstrated high accuracy up to 94.0% in differentiating between expert and novice clinicians. The derived features and trained binary classification models hold promise for improving the accuracy of glaucoma detection and distinguishing between expert and novice ophthalmologists. These findings have implications for enhancing ophthalmic education and for the development of effective diagnostic tools.

Assessing Laparoscopic Surgical Skills of Trainees with Scene Independent and Dependent Eye Gaze Metrics

Eye metrics have been used to objectively evaluate proficiency levels in psychomotor tasks, including aviation, driving, sports, and surgery. Despite extensive research devoted to developing and utilizing eye metrics, the literature does not contain any explicit comparison between scene independent and dependent eye metrics for understanding skill acquisition or characterizing expertise. This study collected eye tracking data from medical students practicing the peg transfer task and computed both scene independent and dependent eye metrics to indicate proficiency. K-means clustering analysis on the eye metrics yielded three clusters corresponding to three proficiency levels which showed significantly different trial completion time. The box plots of the eye-gaze metrics illustrated different patterns of scene independent and dependent eye metrics with respect to proficiency levels, highlighting the need for further examination of these metrics for accurate and useful applications.

Eye movements during motor imagery and execution reveal different visuomotor control strategies in manual interception.

Previous research has investigated the degree of congruency in gaze metrics between action execution (AE) and motor imagery (MI) for similar manual tasks. Although eye movement dynamics seem to be limited to relatively simple actions toward static objects, there is little evidence of how gaze parameters change during imagery as a function of more dynamic spatial and temporal task demands. This study examined the similarities and differences in eye movements during AE and MI for an interception task. Twenty-four students were asked to either mentally simulate or physically intercept a moving target on a computer display. Smooth pursuit, saccades, and response time were compared between the two conditions. The results show that MI was characterized by higher smooth pursuit gain and duration while no meaningful differences were found in the other parameters. The findings indicate that eye movements during imagery are not simply a duplicate of what happens during actual performance. Instead, eye movements appear to vary as a function of the interaction between visuomotor control strategies and task demands.

Mapping models in novel metaphors and their effect on gaze behavior and default interpretations in native and target languages

In the study, we address the problem of existing differences in reading and understanding novel metaphors in the text fragments in native and target languages (L1 and L2), with these differences potentially attributed to both the specifics of forming analogies in native and target languages, and the mapping characteristics of metaphors. The study identifies the contingency effects of several primary metaphors onto the gaze behavior and default interpretation of textual novel metaphors in L1 (Russian) and L2 (English). To proceed, we use the text fragments in L1 and L2 containing novel metaphors appearing in more and less focal syntactic positions in a two-stage oculographic experiment. We obtain the participants’ gaze metrics values and the participants’ responses specifying the target domains of the novel metaphors, which further allows us to disclose the contingencies. Methodologically, the study is grounded in the metaphor processing theories developed in cognitive psychology, which explore the structure of analogical reasoning and associative fluency as manifesting potentially different effects in L1 and L2. To validate it, we also address the cognitive linguistic theories which provide the framework for identifying the primary metaphor models (here the models PATIENT (OBJECT) IS AGENT, PARTS ARE WHOLE, CONCRETE IS ABSTRACT) and for testing their effect onto information construal. We hypothesize that reading and understanding metaphors will proceed differently in L1 and L2, which is attributed to associative fluency in metaphor mapping in native and target languages. The experiment results do not show the differences in understanding the mapping model PATIENT (OBJECT) IS AGENT in L1 and L2, whereas these differences appear in understanding the models PARTS ARE WHOLE and CONCRETE IS ABSTRACT with higher default interpretation index in L1. The model PATIENT (OBJECT) IS AGENT is also found to stimulate higher gaze costs. The results suffice to claim that there are differences in the cognitive costs produced by primary metaphor models, which allows us to range and specify their role in information construal in L1 and L2.

A Platform for Tracking Surgeon and Observer Gaze as a Surrogate for Attention in Ophthalmic Surgery.

To develop and validate a platform that can extract eye gaze metrics from surgeons observing cataract and vitreoretinal procedures and to enable post hoc data analysis to assess potential discrepancies in eye movement behavior according to surgeon experience. Experimental, prospective, single-center study. Eleven ophthalmic surgeons observing deidentified vitreoretinal and cataract surgical procedures performed at a single university-based medical center. An open-source platform was developed to extract gaze coordinates and metrics from ophthalmic surgeons via a computer vision algorithm in conjunction with a neural network to track and segment instruments and tissues, identifying areas of attention in the visual field of study subjects. Eleven surgeons provided validation data by watching videos of 6 heterogeneous vitreoretinal and cataract surgical phases. Accuracy and distance traveled by the eye gaze of participants and overlap of the participants' eye gaze with instruments and tissues while observing surgical procedures. The platform demonstrated repeatability of > 94% when acquiring the eye gaze behavior of subjects. Attending ophthalmic surgeons and clinical fellows exhibited a lower overall cartesian distance traveled in comparison to resident physicians in ophthalmology (P<0.02). Ophthalmology residents and clinical fellows exhibited more fixations to the display area where surgical device parameters were superimposed than attending surgeons (P < 0.05). There was a trend toward gaze overlap with the instrument tooltip location among resident physicians in comparison to attending surgeons and fellows (41.42% vs. 34.8%, P>0.2). The number and duration of fixations did not vary substantially among groups (P>0.3). The platform proved effective in extracting gaze metrics of ophthalmic surgeons. These preliminary data suggest that surgeon gaze behavior differs according to experience.

Eye gaze metrics for skill assessment and feedback in kidney stone surgery.

Surgical skill assessment is essential for safe operations. In endoscopic kidney stone surgery, surgeons must perform a highly skill-dependent mental mapping from the pre-operative scan to the intraoperative endoscope image. Poor mental mapping can lead to incomplete exploration of the kidney and high reoperation rates. Yet there are few objective ways to evaluate competency. We propose to use unobtrusive eye-gaze measurements in the task space to evaluate skill and provide feedback. We capture the surgeons' eye gaze on the surgical monitor with the Microsoft Hololens 2. To enable stable and accurate gaze detection, we develop a calibration algorithm to refine the eye tracking of the Hololens. In addition, we use a QR code to locate the eye gaze on the surgical monitor. We then run a user study with three expert and three novice surgeons. Each surgeon is tasked to locate three needles representing kidney stones in three different kidney phantoms. We find that experts have more focused gaze patterns. They complete the task faster, have smaller total gaze area, and the gaze fewer times outside the area of interest. While fixation to non-fixation ratio did not show significant difference in our findings, tracking the ratio over time shows different patterns between novices and experts. We show that a non-negligible difference holds between novice and expert surgeons' gaze metrics in kidney stone identification in phantoms. Expert surgeons demonstrate more targeted gaze throughout a trial, indicating their higher level of proficiency. To improve the skill acquisition process for novice surgeons, we suggest providing sub-task specific feedback. This approach presents an objective and non-invasive method to assess surgical competence.

Driver eye glance behavior and performance with camera-based visibility systems versus mirrors

Objective Drivers’ ability to extract visual information efficiently from mirrors or camera-based visibility systems impacts driving performance when carrying out maneuvers such as lane changes. The objective of the research was to compare drivers’ eye gaze behavior and driving performance with mirrors versus camera-based visibility systems (i.e., CMS, or camera monitor system) to identify any differences and possible impacts on safety. Methods A test track study was conducted comparing drivers’ eye gaze and lane change behavior when driving a vehicle equipped with outside mirrors versus a prototype CMS. Participants’ opinions regarding usability and comfort in using mirrors versus the tested CMS were also obtained using a post-drive questionnaire. Results Study results were somewhat mixed but did demonstrate that with the tested CMS, participants took longer to pass a slower moving vehicle and maintained a greater resultant distance from the passed vehicle. Additionally, participants had a greater number of fixations to the CMS displays compared to the outside rearview mirrors. Results also found slight perceived advantages for the tested CMS in regard to ease of use, comfortability, and visibility. When asked to choose which rear visibility technology they would prefer to use in everyday driving, most participants preferred the outside rearview mirrors over the tested prototype CMS or having both systems. However, not all lane change and gaze metrics followed the same pattern. Conclusions In this study, participants’ longer time to pass a slower moving vehicle, greater distance when passing a slower moving vehicle, greater number of fixations, and lower subjective ratings with the tested CMS may indicate difficulty in judging distances and focusing on the electronic image. This study provides preliminary findings that suggest differences in driving behavior exist between a single tested prototype CMS and outside rearview mirrors and is a foundational step toward evaluating whether these trends are consistent across different systems and overall implications for safe driving behavior.

Eye Gaze Patterns during Reasoning Provide Insights Regarding Individual Differences in Underlying Cognitive Abilities.

Sequences of eye movements during performance of a reasoning task has provided insights into the strategies individuals use to solve that specific task; however, prior studies have not examined whether eye gaze metrics reflect cognitive abilities in a way that transcends a specific task. Thus, our study aimed to explore the relationship between eye movement sequences and other behavioral measures. Here, we present two studies that related different eye gaze metrics in a matrix reasoning task with performance on a different test of fluid reasoning and tests of planning, working memory, and cognitive flexibility. Additionally, we related gaze metrics to self-reported executive functioning in daily life, as measured by BRIEF-A. To perform this, we classified the participants' eye gaze in each item of the matrices test using an algorithm and then used LASSO regression models with the cognitive abilities as the dependent variable to select eye-tracking metrics to predict it. Overall, specific and distinct eye gaze metrics predicted 57% variance in the fluid reasoning scores; 17% variance in the planning scores; and 18% variance in the working memory scores. Taken together, these results support the hypothesis that the selected eye-tracking metrics reflect cognitive abilities that transcend specific tasks.

Categorical versus dimensional structure of autism spectrum disorder: A multi-method investigation.

A key question for any psychopathological diagnosis is whether the condition is continuous or discontinuous with typical variation. The primary objective of this study was to use a multi-method approach to examine the broad latent categorical versus dimensional structure of autism spectrum disorder (ASD). Data were aggregated across seven independent samples of participants with ASD, other neurodevelopmental disorders (NDD), and non-ASD/NDD controls (aggregate Ns=512-16,755; ages 1.5-22). Scores from four distinct phenotype measures formed composite "indicators" of the latent ASD construct. The primary indicator set included eye gaze metrics from seven distinct social stimulus paradigms. Logistic regressions were used to combine gaze metrics within/across paradigms, and derived predicted probabilities served as indicator values. Secondary indicator sets were constructed from clinical observation and parent-report measures of ASD symptoms. Indicator sets were submitted to taxometric- and latent class analyses. Across all indicator sets and analytic methods, there was strong support for categorical structure corresponding closely to ASD diagnosis. Consistent with notions of substantial phenotypic heterogeneity, the ASD category had a wide range of symptom severity. Despite the examination of a large sample with a wide range of IQs in both genders, males and children with lower IQ were over-represented in the ASD category, similar to observations in diagnosed cases. Our findings provide strong support for categorical structure corresponding closely to ASD diagnosis. The present results bolster the use of well-diagnosed and representative ASD groups within etiologic and clinical research, motivating the ongoing search for major drivers of the ASD phenotype. Despite the categorical structure of ASD, quantitative symptom measurements appear more useful for examining relationships with other factors.

Using eye gaze to reveal cognitive processes and strategies of engineering students when solving spatial rotation and mental cutting tasks

AbstractBackgroundSpatial problem‐solving is an essential skill for success in many engineering disciplines; thus, understanding the cognitive processes involved could help inform the design of training interventions for students trying to improve this skill. Prior research has yet to investigate the differences in cognitive processes between spatial tasks in problem‐solving to offer learners timely feedback.Purpose/HypothesisIn this study, we investigated how different spatial tasks change the cognitive processes and problem‐solving strategies used by engineering students with low spatial ability.Design/MethodStudy participants completed mental rotation and mental cutting tasks of high and low difficulty. Eye‐tracking data were collected and categorized as encoding, transformation, and confirmation cognitive processes. The adoption of either a holistic or piecemeal strategy and response accuracy were also measured.ResultsMental rotation was found to have a higher number of fixations for each cognitive process than the mental cutting task. The holistic strategy was used in both difficulty levels of the mental cutting task, while the piecemeal strategy was adopted for the mental rotation task at a high difficulty level. Only encoding fixations were significantly correlated with accuracy and most strongly correlated with strategy.ConclusionEncoding is an important cognitive process that could affect subsequent cognitive processes and strategies and could, thus, play an important role in performance. Future development in spatial training should consider how to enhance encoding to aid students with low spatial ability. Educators can utilize gaze metrics and empirical research to provide tailored and timely feedback to learners.

Scene-dependent, feedforward eye gaze metrics can differentiate technical skill levels of trainees in laparoscopic surgery.

In laparoscopic surgery, looking in the target areas is an indicator of proficiency. However, gaze behaviors revealing feedforward control (i.e., looking ahead) and their importance have been under-investigated in surgery. This study aims to establish the sensitivity and relative importance of different scene-dependent gaze and motion metrics for estimating trainee proficiency levels in surgical skills. Medical students performed the Fundamentals of Laparoscopic Surgery peg transfer task while recording their gaze on the monitor and tool activities inside the trainer box. Using computer vision and fixation algorithms, five scene-dependent gaze metrics and one tool speed metric were computed for 499 practice trials. Cluster analysis on the six metrics was used to group the trials into different clusters/proficiency levels, and ANOVAs were conducted to test differences between proficiency levels. A Random Forest model was trained to study metric importance at predicting proficiency levels. Three clusters were identified, corresponding to three proficiency levels. The correspondence between the clusters and proficiency levels was confirmed by differences between completion times (F2,488 = 38.94, p < .001). Further, ANOVAs revealed significant differences between the three levels for all six metrics. The Random Forest model predicted proficiency level with 99% out-of-bag accuracy and revealed that scene-dependent gaze metrics reflecting feedforward behaviors were more important for prediction than the ones reflecting feedback behaviors. Scene-dependent gaze metrics revealed skill levels of trainees more precisely than between experts and novices as suggested in the literature. Further, feedforward gaze metrics appeared to be more important than feedback ones at predicting proficiency.

Assessment of changes in gaze patterns during training in point-of-care ultrasound

BackgroundPoint-of-care ultrasound (POCUS) is a core skill in emergency medicine (EM), however, there is a lack of objective competency measures. Eye-tracking technology is a potentially useful assessment tool, as gaze patterns can reliably discriminate between experts and novices across medical specialties. We aim to determine if gaze metrics change in an independent and predictable manner during ultrasound training.MethodsA convenience sample of first-year residents from a single academic emergency department was recruited. Participants interpreted 16 ultrasound videos of the focused assessment with sonography for trauma (FAST) scan while their gaze patterns were recorded using a commercially available eye-tracking device. The intervention group then completed an introductory ultrasound course whereas the control group received no additional education. The gaze assessment was subsequently repeated. The primary outcome was total gaze duration on the area of interest (AOI). Secondary outcomes included time to fixation, mean duration of first fixation and mean number of fixations on the AOI.Results10 EM residents in the intervention group and 10 non-EM residents in the control group completed the study. After training, there was an 8.8 s increase in the total gaze time on the AOI in the intervention group compared to a 4.0 s decrease in the control group (p = .03). EM residents were also 3.8 s quicker to fixate on the AOI whereas the control group became 2.5 s slower (p = .04). There were no significant interactions on the number of fixations (0.43 vs. 0.18, p = .65) or duration of first fixation on the AOI (0.02 s vs. 0.06 s, p = .63).ConclusionsThere are significant and quantifiable changes in gaze metrics, which occur with incremental learning after an ultrasound course. Further research is needed to validate the serial use of eye-tracking technology in following a learner’s progress toward competency in point-of-care ultrasound image interpretation.