Eye-Tracking Technology in Surgical Training

Educational challenges and opportunities for the future generation of surgeons: a scoping review

PURPOSE

Surgery offers exciting opportunities but comes with demanding challenges that require attention from both surgical program administrators and aspiring surgeons. The hashtag #NoTrainingTodayNoSurgeonsTomorrow on 𝕏 (previously Twitter) underscores the importance of ongoing training. Our scoping review identifies educational challenges and opportunities for the next generation of surgeons, analyzing existing studies and filling gaps in the literature.

METHODS

Following the PRISMA guidelines, MEDLINE/PubMed was searched in February 2022, using the MeSH terms "surgeons/education," for articles in English or German on general, abdominal, thoracic, vascular, and hand surgery and traumatology targeting medical students, surgical residents, future surgeons, and fellows.

RESULTS

The initial search yielded 1448 results. After a step-by-step evaluation process, 32 publications remained for complete review. Three main topics emerged: surgical innovations and training (n = 7), surgical culture and environment (n = 19), and mentoring (n = 6). The articles focusing on surgical innovations and training mainly described the incorporation of structured surgical training methods and program initiatives. Articles on surgical culture examined residents' burnout, well-being, and gender issues. Challenges faced by women, including implicit bias and sexual harassment, were highlighted. Regarding mentoring, mentees' needs, training challenges, and the qualities expected of both mentors and mentees were addressed.

CONCLUSION

At a time of COVID-19-driven surgical innovations, the educational and working environment of the new generation of surgeons is changing. Robotic technology and other innovations require future surgeons to acquire additional technological and digital expertise. With regard to the cultural aspects of training, surgery needs to adapt curricula to meet the demands of the new generation of surgeons, but even more it has to transform its culture.

Eye-tracking in surgery: a systematic review

BACKGROUND

Surgery is constantly evolving with the assistance of rapidly developing novel technology. Eye-tracking devices provide opportunities to monitor the acquisition of surgical skills, gain insight into performance, and enhance surgical practice. The aim of this review was to consolidate the available evidence for the use of eye-tracking in the surgical disciplines.

METHODS

A systematic literature review was conducted in accordance with PRISMA guidelines. A search of OVID Medline, EMBASE, Cochrane library, Scopus, and Science Direct was conducted January 2000 until December 2022. Studies involving eye-tracking in surgical training, assessment and technical innovation were included in the review. Non-surgical procedures, animal studies, and studies not involving surgical participants were excluded from the review.

RESULTS

The search returned a total of 12 054 articles, 80 of which were included in the final analysis and review. Seventeen studies involved eye-tracking in surgical training, 48 surgical assessment, and 20 were focussing on technical aspects of this technology. Twenty-six different eye-tracking devices were used in the included studies. Metrics such as the number of fixations, duration of fixations, dwell time, and cognitive workload were able to differentiate between novice and expert performance. Eight studies demonstrated the effectiveness of gaze-training for improving surgical skill.

CONCLUSION

The current literature shows a broad range of utility for a variety of eye-tracking devices in surgery. There remains a lack of standardization for metric parameters and gaze analysis techniques. Further research is required to validate its use to establish reliability and create uniform practices.

Brain computed tomography reading of stroke patients by resident doctors from different medical specialities: An eye-tracking study

BACKGROUND

Using the eye-tracking technique, our work aimed to examine whether difference in clinical background may affect the training outcome of resident doctors' interpretation skills and reading behaviour related to brain computed tomography (CT).

METHODS

Twelve resident doctors in the neurology, radiology, and emergency departments were recruited. Each participant had to read CT images of the brain for two cases. We evaluated each participant's accuracy of lesion identification. We also used the eye-tracking technique to assess reading behaviour. We recorded dwell times, fixation counts, run counts, and first-run dwell times of target lesions to evaluate visual attention. Transition entropy was applied to assess the temporal relations and spatial dynamics of systematic image reading.

RESULTS

The eye-tracking results showed that the image reading sequence examined by transition entropy was comparable among resident doctors from different medical specialties (p = 0.82). However, the dwell time of the target lesions was shorter for the resident doctors from the neurology department (4828.63 ms, p = 0.01) than for those from the resident doctors from the radiology (6275.88 ms) and emergency (5305.00 ms) departments. The eye-tracking results in individual areas of interest only showed differences in the eye-tracking performance of the first-run dwell time (p = 0.05) in the anterior cerebral falx.

DISCUSSION

Our findings demonstrate that resident doctors from different medical specialties may achieve similar imaging reading patterns for brain CT. This may mitigate queries regarding the influence of different backgrounds on training outcomes.

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

Purpose

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.

Design

Experimental, prospective, single-center study.

Participants

Eleven ophthalmic surgeons observing deidentified vitreoretinal and cataract surgical procedures performed at a single university-based medical center.

Methods

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.

Main Outcome Measures

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.

Results

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).

Conclusions

The platform proved effective in extracting gaze metrics of ophthalmic surgeons. These preliminary data suggest that surgeon gaze behavior differs according to experience.

Assessing oral comprehension with an eye tracking based innovative device in critically ill patients and healthy volunteers: a cohort study

PURPOSE

Communication of caregivers and relatives to patients is a major difficulty in intensive care units (ICU). Patient's comprehension capabilities are variable over time and traditional comprehension tests cannot be implemented. Our purpose was to evaluate an oral comprehension test adapted for its automatic implementation using eye-tracking technology among ICU patients.

METHODS

Prospective bi-centric cohort study was conducted on 60 healthy volunteers and 53 ICU patients. Subjects underwent an oral comprehension test using an eye-tracking device: Their results and characteristics were collected. The total duration of the test was 2 and a half minutes.

RESULTS

While performing the test, 48 patients (92%) received invasive ventilation. Among healthy volunteers, the median rate of right answers was very high (93% [interquartile range 87, 100]), whereas it was lower (33% [20, 67]) for patients. For both groups, a significantly lower right answers rate was observed with advancing age (67% [27, 80] vs. 27% [20, 38] among patients and 93% [93, 100] vs. 87% [73, 93] among healthy volunteers, below and above 60 years of age, respectively) and in case of lack of a bachelor's degree (60% [38, 87] vs. 27% [20, 57] among patients and 93% [93, 100] vs. 87% [73, 93] among healthy volunteers). For patients, the higher the severity of disease was, the lower the rate of correct answers was.

CONCLUSION

The eye-tracking-adapted comprehension test is easy and fast to use among ICU patients, and results seem coherent with various potential levels of comprehension as hypothesized in this study.

The Measurement of Cognitive Workload in Surgery Using Pupil Metrics: A Systematic Review and Narrative Analysis

INTRODUCTION

Increased cognitive workload (CWL) is a well-established entity that can impair surgical performance and increase the likelihood of surgical error. The use of pupil and gaze tracking data is increasingly being used to measure CWL objectively in surgery. The aim of this review is to summarize and synthesize the existing evidence that surrounds this.

METHODS

A systematic review was undertaken in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A search of OVID MEDLINE, IEEE Xplore, Web of Science, Google Scholar, APA PsychINFO, and EMBASE was conducted for articles published in English between 1990 and January 2021. In total, 6791 articles were screened and 32 full-text articles were selected based on the inclusion criteria. A narrative analysis was undertaken in view of the heterogeneity of studies.

RESULTS

Seventy-eight percent of selected studies were deemed high quality. The most frequent surgical environment and task studied was surgical simulation (75%) and performance of laparoscopic skills (56%) respectively. The results demonstrated that the current literature can be broadly categorized into pupil, blink, and gaze metrics used in the assessment of CWL. These can be further categorized according to their use in the context of CWL: (1) direct measurement of CWL (n = 16), (2) determination of expertise level (n = 14), and (3) predictors of performance (n = 2).

CONCLUSIONS

Eye-tracking data provide a wealth of information; however, there is marked study heterogeneity. Pupil diameter and gaze entropy demonstrate promise in CWL assessment. Future work will entail the use of artificial intelligence in the form of deep learning and the use of a multisensor platform to accurately measure CWL.

Eye Tracking Use in Surgical Research: A Systematic Review

INTRODUCTION

Eye tracking (ET) is a popular tool to study what factors affect the visual behaviour of surgical team members. To our knowledge, there have been no reviews to date that evaluate the broad use of ET in surgical research. This review aims to identify and assess the quality of this evidence, to synthesize how ET can be used to inform surgical practice, and to provide recommendations to improve future ET surgical studies.

METHODS

In line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, a systematic literature review was conducted. An electronic search was performed in MEDLINE, Cochrane Central, Embase, and Web of Science databases up to September 2020. Included studies used ET to measure the visual behaviour of members of the surgical team during surgery or surgical tasks. The included studies were assessed by two independent reviewers.

RESULTS

A total of 7614 studies were identified, and 111 were included for data extraction. Eleven applications were identified; the four most common were skill assessment (41%), visual attention assessment (22%), workload measurement (17%), and skills training (10%). A summary was provided of the various ways ET could be used to inform surgical practice, and three areas were identified for the improvement of future ET studies in surgery.

CONCLUSIONS

This review provided a comprehensive summary of the various applications of ET in surgery and how ET could be used to inform surgical practice, including how to use ET to improve surgical education. The information provided in this review can also aid in the design and conduct of future ET surgical studies.

When medical trainees encountering a performance difficulty: evidence from pupillary responses

BACKGROUND

Medical trainees are required to learn many procedures following instructions to improve their skills. This study aims to investigate the pupillary response of trainees when they encounter moment of performance difficulty (MPD) during skill learning. Detecting the moment of performance difficulty is essential for educators to assist trainees when they need it.

METHODS

Eye motions were recorded while trainees practiced the thoracostomy procedure in the simulation model. To make pupillary data comparable among trainees, we proposed the adjusted pupil size (APS) normalizing pupil dilation for each trainee in their entire procedure. APS variables including APS, maxAPS, minAPS, meanAPS, medianAPS, and max interval indices were compared between easy and difficult subtasks; the APSs were compared among the three different performance situations, the moment of normal performance (MNP), MPD, and moment of seeking help (MSH).

RESULTS

The mixed ANOVA revealed that the adjusted pupil size variables, such as the maxAPS, the minAPS, the meanAPS, and the medianAPS, had significant differences between performance situations. Compared to MPD and MNP, pupil size was reduced during MSH. Trainees displayed a smaller accumulative frequency of APS during difficult subtask when compared to easy subtasks.

CONCLUSIONS

Results from this project suggest that pupil responses can be a good behavioral indicator. This study is a part of our research aiming to create an artificial intelligent system for medical trainees with automatic detection of their performance difficulty and delivering instructional messages using augmented reality technology.

Utilizing head-mounted eye trackers to analyze patterns and decision-making strategies of 3D virtual modelling platform (IRIS™) during preoperative planning for renal cancer surgeries

PURPOSE

IRIS^™ provides interactive, 3D anatomical visualizations of renal anatomy for pre-operative planning that can be manipulated by altering transparency, rotating, zooming, panning, and overlaying the CT scan. Our objective was to analyze how eye tracking metrics and utilization patterns differ between preoperative surgical planning of renal masses using IRIS and CT scans.

METHODS

Seven surgeons randomly reviewed IRIS and CT images of 9 patients with renal masses [5 high complexity (RENAL score ≥ 8), 4 low complexity (≤ 7)]. Surgeons answered a series of questions regarding patient anatomy, perceived difficulty (/100), confidence (/100), and surgical plan. Eye tracking metrics (mean pupil diameter, number of fixations, and gaze duration) were collected.

RESULTS

Surgeons spent significantly less time interpreting data from IRIS than CT scans (- 67.1 s, p < 0.01) and had higher inter-rater agreement of surgical approach after viewing IRIS (α = 0.16-0.34). After viewing IRIS, surgical plans although not statistically significant demonstrated a greater tendency towards a more selective ischemia approaches which positively correlated with improved identification of vascular anatomy. Planned surgical approach changed in 22/59 of the cases. Compared to viewing the CT scan, left and right mean pupil diameter and number/duration of fixations were significantly lower when using IRIS (p < 0.01, p < 0.01, p = 0.42, p < 0.01, respectively), indicating interpreting information from IRIS required less mental effort despite under-utilizing its interactive features.

CONCLUSIONS

Surgeons extrapolated more detailed information in less time with less mental effort using IRIS than CT scans and proposed surgical approaches with potential to enhanced surgical outcomes.

Eye-Tracking Indicators of Workload in Surgery: A Systematic Review

BackgroundEye tracking is a powerful tool for unobtrusive and real time assessment of workload in clinical settings. Before the complex eye tracking derived surrogates can be proactively utilized to improve surgical safety, the indications, validity and reliability requires careful evaluation.MethodsWe conducted a systematic review of literature from 2010 to 2020 according to PRISMA guidelines. A search on PubMed, Cochrane, Scopus, Web of science, PsycInfo and Google scholar databases was conducted on July 2020. The following search query was used" ("eye tracking" OR "gaze tracking") AND (surgery OR surgical OR operative OR intraoperative) AND (workload OR stress)". Short papers, no peer reviewed or papers in which eye-tracking methodology was not used to investigate workload or stress factors in surgery, were omitted.ResultsA total of 17 (N = 17) studies were identified eligible to this review. Most of the studies (n = 15) measured workload in simulated setting. Task difficulty and expertise were the most studied factors. Studies consistently showed surgeon's eye movements such as pupil responses, gaze patterns, blinks were associated with the level of perceived workload. However, differences between measurements in operational room and simulated environments have been found.ConclusionPupil responses, blink rate and gaze indices are valid indicators of workload. However, the effect of distractions and non-technical factors on workload is underrepresented aspect in the literature even though recognized as underlying factors in successful surgery.

Use of Artificial Intelligence in Non-Oncologic Interventional Radiology: Current State and Future Directions

The future of radiology is disproportionately linked to the applications of artificial intelligence (AI). Recent exponential advancements in AI are already beginning to augment the clinical practice of radiology. Driven by a paucity of review articles in the area, this article aims to discuss applications of AI in non-oncologic IR across procedural planning, execution, and follow-up along with a discussion on the future directions of the field. Applications in vascular imaging, radiomics, touchless software interactions, robotics, natural language processing, post-procedural outcome prediction, device navigation, and image acquisition are included. Familiarity with AI study analysis will help open the current 'black box' of AI research and help bridge the gap between the research laboratory and clinical practice.

How Wearable Technology Can Facilitate AI Analysis of Surgical Videos

Operative video has great potential to enable instant replays of critical surgical decisions for training and quality review. Recently, artificial intelligence (AI) has shown early promise as a method of enabling efficient video review, analysis, and segmentation. Despite the progress with AI analysis of surgical videos, more work needs to be done to improve the accuracy and efficiency of AI-driven video analysis. At a recent consensus conference held on July 10-11, 2020, 8 research teams shared their work using AI for surgical video analysis. Four of the teams showcased the utility of wearable technology in providing objective surgical metrics. Data from these technologies were shown to pinpoint important cognitive and motor actions during operative tasks and procedures. The results support the utility of wearable technology to facilitate efficient and accurate video analysis and segmentation.

Modeling Intensive Polytomous Time-Series Eye-Tracking Data: A Dynamic Tree-Based Item Response Model

This paper presents a dynamic tree-based item response (IRTree) model as a novel extension of the autoregressive generalized linear mixed effect model (dynamic GLMM). We illustrate the unique utility of the dynamic IRTree model in its capability of modeling differentiated processes indicated by intensive polytomous time-series eye-tracking data. The dynamic IRTree was inspired by but is distinct from the dynamic GLMM which was previously presented by Cho, Brown-Schmidt, and Lee (Psychometrika 83(3):751-771, 2018). Unlike the dynamic IRTree, the dynamic GLMM is suitable for modeling intensive binary time-series eye-tracking data to identify visual attention to a single interest area over all other possible fixation locations. The dynamic IRTree model is a general modeling framework which can be used to model change processes (trend and autocorrelation) and which allows for decomposing data into various sources of heterogeneity. The dynamic IRTree model was illustrated using an experimental study that employed the visual-world eye-tracking technique. The results of a simulation study showed that parameter recovery of the model was satisfactory and that ignoring trend and autoregressive effects resulted in biased estimates of experimental condition effects in the same conditions found in the empirical study.