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

Advancements in microsurgery: A comprehensive systematic review of artificial intelligence applications

Machine learning (ML) is a branch of artificial intelligence (AI) that enables computers to learn from data and discern patterns without direct instruction. This review explores cutting-edge developments in microsurgery through the lens of AI applications. By analyzing a wide range of studies, this paper highlights AI's transformative role in enhancing microsurgical techniques and decision-making processes. A systematic literature search was conducted using Ovid MEDLINE, Ovid Embase, Web of Science, and PubMed (2005-2023). Extensive data on ML model function and composition, as well as broader study characteristics, were collected from each study. Study quality was assessed across 7 methodological areas of AI research using an adapted methodological index of nonrandomized studies (MINORS) tool. Seventeen studies met the inclusion criteria. ML was used primarily for prognosis (35%), postoperative assessment (29%), and intraoperative assistance/robotic surgery (24%). Only 2 studies were conducted beyond phase 0 of AI research. Fourteen studies included a training group, but only one of these reported both validation and training sets. ML model performance was assessed most frequently using accuracy, specificity, and sensitivity. Scores for the adapted MINORS criteria ranged from 10 to 14 out of 14, with a median of 12. Through collation of all available preclinical and clinical trials, this review suggests the efficacy of ML for various microsurgical applications. Despite this, widespread adoption of this technology remains scarce, currently limited by methodological flaws of individual studies and structural barriers to disruptive technologies. However, with growing evidence supporting its use, microsurgeons should be receptive to implementing ML-incorporated technologies or may risk falling behind other specialties.

Interventions to prevent visual fatigue during robotic surgery

The robotic surgeon is at risk of visual fatigue from prolonged viewing of the video display resulting in digital eye strain and use of the three-dimensional binoculars resulting in accommodative stress. Symptoms of digital eye strain include blurred vision, dry eyes, eyestrain, neck and back ache, diplopia, light sensitivity, and headaches. Vergence or accommodation-related symptoms include blurred near or distance vision, difficulty refocusing, and diplopia. Beneficial ergonomic interventions to manage digital eye strain during robotic surgery include appropriate lighting, improved neck positioning, optimal screen positioning, improved image parameters, screen breaks, optimising environmental factors, and eye exercises. Correction of refractive error, use of lubricating eye drops, and blink efficiency training to induce motor memory have been shown to be effective in reducing visual fatigue. Vergence-accommodation mismatch can be reduced with slower movement of the camera, screen breaks, and correction of refractive error. Robotic surgeons should adopt these simple and non-invasive interventions to minimise visual fatigue.

Novel multimodal sensing and machine learning strategies to classify cognitive workload in laparoscopic surgery

BACKGROUND

Surgeons can experience elevated cognitive workload (CWL) during surgery due to various factors including operative technicalities and the environmental demands of the operating theatre. This can result in poorer outcomes and have a detrimental effect on surgeon well-being. The objective measurement of CWL provides a potential solution to facilitate classification of workload levels, however results are variable when physiological measures are used in isolation. The aim of this study is to develop and propose a multimodal machine learning (ML) approach to classify CWL levels using a bespoke sensor platform and to develop a ML approach to impute missing pupil diameter measures due to the effect of blinking or noise.

MATERIALS AND METHODS

Ten surgical trainees performed a simulated laparoscopic cholecystectomy under cognitive conditions of increasing difficulty, namely a modified auditory N-back task with increasing difficulty and a verbal clinical scenario. Physiological measures were recorded using a novel platform (MAESTRO). Electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) were used as direct measures of CWL. Indirect measures included electromyography (EMG), electrocardiography (ECG) and pupil diameter (PD). A reference point for validation was provided by subjective assessment of perceived CWL using the SURG-TLX. A multimodal machine learning approach that systematically implements a CNN-BiLSTM, a binary version of the metaheuristic Manta Ray Foraging Optimisation (BMRFO) and a version of Fuzzy C-Means (FCM) called Optimal Completion Strategy (OCS) was used to classify the associated perceived CWL state.

RESULTS

Compared to other state of the art classification techniques, cross-validation results for the classification of CWL levels suggest that the CNN-BLSTM and BMRFO approach provides an average accuracy of 97 % based on the confusion matrix. Additionally, OCS demonstrated a superior average performance of 9.15 % in terms of Root-Mean-Square-Error (RMSE) when compared to other PD imputation methods.

CONCLUSION

Perceived CWL levels were correctly classified using a multimodal ML approach. This approach provides a potential route to accurately classify CWL levels, which may have application in future surgical training and assessment programs as well as the development of cognitive support systems in the operating room.

Pupil Response in Visual Tracking Tasks: The Impacts of Task Load, Familiarity, and Gaze Position

Pupil size is a significant biosignal for human behavior monitoring and can reveal much underlying information. This study explored the effects of task load, task familiarity, and gaze position on pupil response during learning a visual tracking task. We hypothesized that pupil size would increase with task load, up to a certain level before decreasing, decrease with task familiarity, and increase more when focusing on areas preceding the target than other areas. Fifteen participants were recruited for an arrow tracking learning task with incremental task load. Pupil size data were collected using a Tobii Pro Nano eye tracker. A 2 × 3 × 5 three-way factorial repeated measures ANOVA was conducted using R (version 4.2.1) to evaluate the main and interactive effects of key variables on adjusted pupil size. The association between individuals' cognitive load, assessed by NASA-TLX, and pupil size was further analyzed using a linear mixed-effect model. We found that task repetition resulted in a reduction in pupil size; however, this effect was found to diminish as the task load increased. The main effect of task load approached statistical significance, but different trends were observed in trial 1 and trial 2. No significant difference in pupil size was detected among the three gaze positions. The relationship between pupil size and cognitive load overall followed an inverted U curve. Our study showed how pupil size changes as a function of task load, task familiarity, and gaze scanning. This finding provides sensory evidence that could improve educational outcomes.

Cognitive ergonomics and robotic surgery

Cognitive ergonomics refer to mental resources and is associated with memory, sensory motor response, and perception. Cognitive workload (CWL) involves use of working memory (mental strain and effort) to complete a task. The three types of cognitive loads have been divided into intrinsic (dependent on complexity and expertise), extraneous (the presentation of tasks) and germane (the learning process) components. The effect of robotic surgery on CWL is complex because the postural, visualisation, and manipulation ergonomic benefits for the surgeon may be offset by the disadvantages associated with team separation and reduced situation awareness. Physical fatigue and workflow disruptions have a negative impact on CWL. Intraoperative CWL can be measured subjectively post hoc with the use of self-reported instruments or objectively with real-time physiological response metrics. Cognitive training can play a crucial role in the process of skill acquisition during the three stages of motor learning: from cognitive to integrative and then to autonomous. Mentorship, technical practice and watching videos are the most common traditional cognitive training methods in surgery. Cognitive training can also occur with computer-based cognitive simulation, mental rehearsal, and cognitive task analysis. Assessment of cognitive skills may offer a more effective way to differentiate robotic expertise level than automated performance (tool-based) metrics.

Examining workload variations among different surgical team roles, specialties, and techniques: a multicenter cross-sectional descriptive study

BACKGROUND

A high workload may negatively impact the surgical team's performance and jeopardize patient safety. The aim of this study was to measure the workload of the surgical team across different surgical roles, specialties, and techniques in several hospitals.

METHODS

This cross-sectional multicenter study was performed in the operating rooms of eight teaching hospitals affiliated with Isfahan University of Medical Sciences, Iran. At the conclusion of each surgical procedure, all members of the surgical team completed the Surgery Task Load Index (SURG-TLX) questionnaire to assess workload levels. Descriptive statistics, analysis of variance (ANOVA), and Pearson correlations, were performed to compare surgical roles, specialties, techniques, and surgical time on workload overall and by subscale.

RESULTS

A total of 409 workload questionnaires were obtained from 76 surgical teams or cases, involving 346 surgical team members. The total workload among all participants was 32.41 ± 17.21. Surgical complexity, physical demands, and mental demands were the highest workload subscales and distraction was the lowest workload subscale. Cardiovascular specialty had a higher workload compared to other specialties. Open techniques resulted in a higher workload compared to minimally invasive techniques. Surgical technologists who act in both the role of circulating and scrub nurse (C&Ss) experienced the highest workload, followed by surgical residents and surgeons.

CONCLUSIONS

The results of the study showed that the workload for some members of the surgical team is disproportionately high and is influenced by factors such as specialty, technique, role, and surgical duration. By knowing the distribution of workload among the members of the surgical team, efforts can be made to optimize the team members' workload.

Expertise differences in cognitive interpreting: A meta-analysis of eye tracking studies across four decades

This meta-analytic research delves into the influence of expertise on cognitive interpreting, emphasizing time efficiency, accuracy, and cognitive effort, in alignment with prevailing expertise theories that link professional development and cognitive efficiency. The study assimilates empirical data from 18 eye-tracking studies conducted over the past four decades, encompassing a sample of 1581 interpreters. The objective is to elucidate the role of expertise in interpretative performance while tracing the evolution of these dynamics over time. Findings suggest that expert interpreters outperform novices in time efficiency and accuracy and exhibit lower cognitive effort, especially in sight and consecutive interpreting. This effect is particularly pronounced in the English-Chinese language pair and with the use of E-prime and Tobii eye-tracking systems. Further, fixation count and pupil size are essential metrics impacting cognitive effort. These findings have vital implications for interpreter training programs, suggesting a focus on expertise development to enhance efficiency and accuracy, reduce cognitive load, and emphasize the importance of sight interpreting as a foundational skill. The selection of technology and understanding of specific ocular metrics also emerged as essential for future research and practical applications in the interpreting industry. This article is categorized under: Psychology > Theory and Methods Linguistics > Cognitive.

Cognitive Ergonomics: A Review of Interventions for Outpatient Practice

Doctoring is difficult mental work, involving many cognitively demanding processes such as diagnosing, decision-making, parallel processing, communicating, and managing the emotions of others. According to cognitive load theory (CLT), working memory is a limited cognitive resource that can support a finite amount of cognitive load. While the intrinsic cognitive load is the innate load associated with a task, the extraneous load is generated by inefficiency or suboptimal work conditions. Causes of extraneous cognitive load in healthcare include inefficiency, distractions, interruptions, multitasking, stress, poor communication, conflict, and incivility. High levels of cognitive load are associated with impaired function and an increased risk of burnout among physicians. Cognitive ergonomics is the branch of human factors and ergonomics (HFE) focused on supporting the cognitive processes of individuals within a system. In health care, where the cognitive burden on physicians is high, cognitive ergonomics can establish practices and systems that decrease extraneous cognitive load and support pertinent cognitive processes. In this review, we present cognitive ergonomics as a useful framework for conceptualizing an oft-overlooked dimension of labor and apply theory to practice by summarizing evidence-based cognitive ergonomics interventions for outpatient care settings. Our proposed interventions are structured within four general recommendations: 1. minimize distractions, interruptions, and multitasking; 2. optimize the use of the electronic health record (EHR); 3. optimize the use of health information systems (HIS); and 4. support good communication and teamwork. Best practices in cognitive ergonomics can benefit patients, minimize practice inefficiency, and support physician career longevity.

The Pupil Knows: Pupil Dilation Indexes and Their Inhibitory Ability in Normal Aging

Pupil dilation is considered an index of cognitive effort, as the pupil typically dilates as the cognitive load increases. In this paper, we evaluated whether older adults demonstrate increased pupil size when performing tasks requiring cognitive inhibition. We invited 44 older and 44 younger adults to perform the Stroop task while their pupil dilation was recorded with eye-tracking glasses. The dependent variables were the number of accurate responses on the Stroop task as well as pupil size in the three conditions of the task (i.e., color naming, word reading, and the interference condition). The results demonstrated less accurate responses in the interference condition than in the color-naming or word-reading conditions, in both older and younger adults. Critically, larger pupil dilation was observed in the interference condition than in the color-naming and word-reading conditions, in both older and younger adults. This study demonstrates that pupil dilation responds to cognitive effort in normal aging, at least in the interference condition of the Stroop task.

Transformer Based Cross-Subject Mental Workload Classification Using FNIRS for Real-World Application

Mental state monitoring is a hot topic especially in neurorehabilitation, skill training, etc, for which the functional near-infrared spectroscopy (fNIRS) has been suggested to be used, and fewer detection channels and cross-subject performance are usually required for real-world application. To this goal, we propose a transformer-based method for cross-subject mental workload classification using fewer channels of fNIRS. Firstly, the input fNIRS signals in a window are divided into patches in the temporal order and transformed into embeddings, to which a classification token and learnable position embeddings are added. Then, a transformer encoder is used to learn the long-range dependencies among the embeddings, of which the output classification token is sent to a multilayer perceptron (MLP) head. Mental workload classification results can be represented by the outputs of the MLP head. Finally, comparison experiments were conducted on the open-access fNIRS2MW dataset. The results show that, the proposed method can outperform previous methods in cross-subject classification accuracy, and relatively efficient computation can be obtained.

A Systematic Review of In-Vehicle Physiological Indices and Sensor Technology for Driver Mental Workload Monitoring

The concept of vehicle automation ceases to seem futuristic with the current advancement of the automotive industry. With the introduction of conditional automated vehicles, drivers are no longer expected to focus only on driving activities but are still required to stay alert to resume control. However, fluctuations in driving demands are known to alter the driver's mental workload (MWL), which might affect the driver's vehicle take-over capabilities. Driver mental workload can be specified as the driver's capacity for information processing for task performance. This paper summarizes the literature that relates to analysing driver mental workload through various in-vehicle physiological sensors focusing on cardiovascular and respiratory measures. The review highlights the type of study, hardware, method of analysis, test variable, and results of studies that have used physiological indices for MWL analysis in the automotive context.