Electroencephalographic Monitoring of Brain Wave Activity During Laparoscopic Surgical Simulation to Measure Surgeon Concentration and Stress: Can the Student Become the Master?

Introduction

Historical methods of surgical training and assessment have evolved from “See one, Do one, Teach one” to the current Accreditation Council for Graduate Medical Education Milestone Project that has improved evaluation of physician competency in achieving stratified resident performance levels. ¹ Contemporary final assessment of residents remains subjective, however, relying on the sentiment of the reviewer's opinion and performance to determine the accomplishment of targeted goals.

Although performance metrics are a mainstay of surgical education, the adoption of new techniques requires that the methods in which proficiency are evaluated must evolve concurrently. Minimally invasive surgery (MIS), while offering many advantages to patients, presented unique challenges to both experienced and proficient surgeons as well as novice surgeons or resident trainees. As a result, MIS ushered in an era of renewed interest in understanding the factors affecting surgical learning and a need for tools to measure progress along the process of surgical education (the learning curve).

Traditional markers of trainee laparoscopic skills include subjective assessment of cognitive skills, surgical case logs, and summative evaluations by residency program directors. ² Surgical performance, however, is a dynamic entity dependent on a complex interaction of sensation, perception, cognition, and kinesthetic. Because many of these factors are not readily observed objectively, the literature dedicated to metrics of surgical education is largely focused on the kinesthetic aspect of surgery.

Despite the challenges in quantitating cognitive skills, they remain an important component of surgical competence, and although surgical simulation and measuring objective surgical performance are in their relative infancy, there is good evidence to suggest that stress can negatively influence performance in other “high-stake” careers such as aviation, military, and competitive sports. ^{3 –6} Because of stress, impaired psychomotor performance may be specifically and especially pronounced in novice laparoscopic surgeons. ⁷

Electroencephalography (EEG) is the measurement of electric activity produced by the brain. Gamma waves represent neurons coming together in a network for carrying out a cognitive or motor function. ⁸ It represents the ability to concentrate and focus. Alpha wave activity is most closely linked to stress or mental exertion. ⁹

We used a wireless, noninvasive EEG brain monitor (MUSE^™ headband) as an objective measurement of concentration and stress during laparoscopic skill tasks on the EDGE simulator. We used tasks from the basic laparoscopic urologic surgery (BLUS) curriculum of the American Urological Association. We previously determined that task completion times and accuracy correlated with participants' laparoscopic surgical experience. ¹⁰ In the present study, we postulate that along with the kinesthetic advantage provided by increased laparoscopic surgical experience, the more experienced participants would also demonstrate decreased stress levels and increased concentration during task performance compared with less experienced participants.

Materials and Methods

After obtaining approval by the Institutional Review Board of Tulane University, participants from the Department of Urology and the School of Medicine were recruited for study enrollment. Participants consisted of three groups spanning a novice, intermediate, and expert level of laparoscopic proficiency based on laparoscopic surgical experience. The novice group was composed of 1st and 2nd year medical student volunteers with no previous surgical experience. The intermediate group consisted of urology residents in postgraduate years 2 to 5 of resident training. The expert group consisted of urology department attending physicians with both extensive previous laparoscopic experience as well as an active practice that includes an MIS focus.

Two of the four validated BLUS tasks were used in the assessment of participant stress and concentration: peg transfer and laparoscopic suturing. Each task was explained and demonstrated for each participant before subject testing by the study administrator. Each task was completed on the EDGE laparoscopic simulator (SimuLab; Seattle, WA) as described previously. ¹⁰

MUSE is a commercially available wireless EEG monitor, which is worn as a flexible, adjustable, lightweight headband with seven sensors: two forehead sensors, two ear sensors, and three reference sensors. The headset is capable of reading alpha, beta, theta, and delta brain waves. For our study, we targeted gamma waves, which are in the frequency range of 30 to 100 Hz, as well as alpha waves in the frequency range of 7 to 14 Hz. The seven-sensor montage enables estimation of hemispheric asymmetries (with an emphasis on the prefrontal lobe) and thus facilitates brain state discrimination that depends on asymmetry, such as emotional valence (positive vs negative emotions). The headband incorporates a three-axis accelerometer enabling quantification of head movements. The simulator and EEG headband are shown in Figure 1. With these features, it allowed us to conduct our specific study.

OPEN IN VIEWER

FIG. 1.

Participant with MUSE^™ headband performing task on the EDGE simulator.

While participants performed the two tasks, we measured concentration and stress levels, using the software development kit (SDK) provided by MUSE. The SDK allows the capability of collecting various types of data through multiple applications, but for our study, we tailored the functionality of the program to extract specifically gamma and alpha wave data. After data collection, we quantified the concentration and stress levels of each participant for each task. All data were collected using MUSE and analyzed after the study was completed. Participants wore the headset throughout the study while data was transmitted via Bluetooth to the personal computer.

Participants performed each of the two BLUS tasks twice, and task completion times (TCTs) were recorded. A continuous EEG tracing for both gamma and alpha waves was collected for each participant and task, as shown in Figure 2. The area under the curve (AUC) was calculated for the gamma and alpha wave tracings. Because TCTs varied significantly among participants, the AUC was divided by time to give a composite concentration and stress score. Statistical analysis consisted of an analysis of variance to compare differences in gamma and alpha EEG wave activity between groups. A critical two-sided P value of <0.05 was used for significance.

OPEN IN VIEWER

FIG. 2.

(A) Representative medical student electroencephalography (EEG) tracing. (B) Representative expert EEG tracing.

Results

A total of 19 participants were recruited into the study with novice, intermediate, and expert laparoscopy experience groups comprised of 6 medical students, 9 urology residents, and 4 attending urologists, respectively. The mean peg transfer TCT was 110, 136, and 205 seconds for the expert, intermediate, and novice groups, respectively (P = 0.00035). The mean suturing TCT was 107, 195, and 279 seconds for the expert, intermediate, and novice groups, respectively (P = 0.000082). The task completion times are shown in Figure 3.

OPEN IN VIEWER

FIG. 3.

Peg transfer and suturing task completion times.

Because alpha activity reflects a relative state of calm, the stress composite score is inversely related to the stress level. Conversely, gamma activity represents a concentrated state, and thus the concentration score is directly proportional to the concentration level of the participant. The composite stress and concentration scores of the different groups for both tasks are demonstrated in Figure 4.

OPEN IN VIEWER

FIG. 4.

Stress and concentration scores for (A) peg transfer and (B) suturing.

Stress was significantly lower in the attending urologists compared with the residents and medical students during the laparoscopic suturing and trended toward significance in the peg transfer task (P = 0.0003, P = 0.06). Concentration levels were significantly better in the expert group compared with the less experienced groups during both the peg and suture tasks (P = 0.036, P = 0.0039).

Discussion

As a direct result of new technologies in surgery, most notably being the development of laparoscopy, there has been a concurrent change in the philosophy of resident surgical education. With this change, an emphasis on surgical simulation has increased, resulting in a departure from the Halstedian principles of surgery training paradigm that relied primarily on learning from repetition of clinical exposures. ¹¹ Because of the lack of clinical outcomes with simulative tasks, educators rely more heavily on surrogate markers of surgical competence.

Electromagnetic motion (EM) tracking systems have been used widely in many different fields, but only recently has this method been applied to surgery. Using EM technology to track hand motion analysis (HMA: total number of movements, movement velocity, frequency, trajectory, and hand travel distance), Datta and associates ¹² correlated technical skill with HMA efficiency in a laboratory setting. HMA was later applied in a live patient setting with similar results, showing HMA's ability to discriminate novice from experienced surgeon's technical skills during vasectomy and vasectomy reversals. ¹³

EM tracking requires the use of EM sensors that must be attached to the surgeon's hands. This makes EM technology somewhat cumbersome and potentially impractical in a sterile operative setting. Limitations of EM systems lead to the introduction of video-based motion analysis (VMA), which obviates the need for bulky sensors and any other cumbersome and potentially expensive element of the EM tracking system. Using VMA, Glarner and colleagues ¹⁴ were able to distinguish experienced from novice surgeons during reduction mammoplasties based on hand kinematics.

There is no debate that technical skills and the acquisition of these is an essential component of surgery and surgical training. There are many less tangible and nontechnical skills, however, that are significant constituents of surgical competence such as communication skills, concentration, and stress management. ¹⁵ Although the literature on methodologies for measuring surgical proficiency is growing, data on more often subjectively measured variables such as stress and concentration are less evolved.

Surrogate markers of concentration have historically centered on eye movements. Electrooculogram has been used to measure eye blink rate, which was significantly decreased indicating concentration on both an open and laparoscopic simulated task compared with a resting state. ¹⁶ Similarly, Richstone and coworkers ¹⁷ demonstrated the relationship between eye metrics and visual attention, insight, and surgical proficiency.

A stress response can occur in two forms: A perceptual or physical response. Perceptual stress is most commonly measured exercising self-reported questionnaires. ¹⁸ A number of surrogates, however, have been described to measure mental stress. These markers of stress include salivary cortisol, heart rate variability, and skin conductance levels, which have been reproducibly altered in response to stressful situations. ^7,16,19 All of these tests are expensive, cumbersome, and are indirect measurements of mental stress by measuring downstream activity of the central autonomic nervous system and endocrine systems, respectively.

In our study, we used the MUSE headband with its EEG readings as a direct measure of cortical activity during laparoscopic simulations. This is in stark comparison to previously mentioned surrogate measures that are not directly measuring brain activity but rather are downstream extrapolations. From previous work, we know that gamma waves are associated with concentration, thought, and listening, while alpha waves reflect a state of calm and balanced psychological state. ²⁰ We found that participants with more experience and thus familiarity with laparoscopy demonstrated significantly less stress based on alpha wave activity for the suturing task, while stress levels trended toward significance during the peg transfer task. We also demonstrated an increase in gamma wave activity indicating an improved state of attentiveness in the expert group compared with the less experienced participants. These EEG observations correlated with faster task completion times, indicating a potential for the EEG findings to correlate with task proficiency.

The purpose of our study was to test whether EEG measurements could represent a tool to measure surgeon stress and concentration levels. An added benefit of the headband tool is its ability to not only measure these variables, but also allow an opportunity for improvement through neurofeedback training (NFT) provided by the original consumer application of MUSE. NFT provides a moment-to-moment feedback to the subject regarding their physiologic functioning, particularly the function of the brain and central nervous system. This uses an operant conditioning paradigm in which participants learn to influence their brains' electrical activity. ²¹ Neurofeedback has previously demonstrated the capacity to improve attention. ^22,23

This provides a unique ability of the MUSE headband to not only be descriptive of surgeon's stress and concentration, but it also may be a therapeutic tool in which the operator can modulate response with the goal of maximizing attentiveness and decreasing stress. In addition, because the device also contains accelerometers, it could potentially be used to perform studies of surgeon ergonomics, neck strain, and movement during surgery. With further study, it may also provide a means of measuring both technical and cognitive progress along the learning curve of surgical learning.

There are some limitations of our study that deserve mention. Our study population was limited, and our intent is to include more subjects as well as potentially perform a multi-institutional study in future validation of the MUSE construct. We believe our data are a direct and objective measurement of concentration and stress, but we did not perform a subjective questionnaire self-evaluation for perceptual correlation. This will be pursued in future study designs. These findings give a quantitative feedback to the students, which may help in decreasing the learning curve. The assessment may help in understanding currently where they are on the learning curve, and over time whether they have become more comfortable and lowered their stress level during the procedure.

Finally, we only performed two task repeats in one setting, which leaves any ability of the headband to be assessed as a NFT tool impossible. Moving forward, we plan to perform laparoscopic assessments on a longitudinal basis in which progress and changes in task proficiency, stress, and concentration levels can be compared and correlated. Even with high fidelity simulators, the translatability of simulation data to an actual situation is not perfect. In addition the EEG headband will be used during live surgery in future testing in which unforeseen factors may influence concentration and stress levels, which may not be readily evident in a simulator setting.

Conclusions

The MUSE headband measured EEG activity during laparoscopic simulation, and its SDK yielded data to specific brain waves related to surgeon concentration and stress levels. As expected, surgeon experience and familiarity correlated well with the optimal cognitive setting of increased concentration coupled with decreased stress, which lead to improved performance.