Non-invasive brain imaging of laparoscopic motor skills learning

Arun Nemani, PhD¹, Clairice Cooper, MD², Steven Schwaitzberg, MD², Suvranu De, MD¹. ¹Rensselaer Polytechnic Institute, ²University of Buffalo

INTRODUCTION: Research has clearly shown the benefits of surgical simulators to help train laparoscopic motor skills required for positive patient outcomes. Although there are many studies that research the performance metrics of motor skills, very few studies have researched the cortical activation changes that undergo during the various stages of surgical motor skill learning. We propose a non-invasive brain imaging technique, called functional near-infrared spectroscopy (fNIRS) to measure cortical activity in the prefrontal cortex (PFC), primary motor cortex (M1) and the supplementary motor area (SMA) as subjects attain laparoscopic skills mastery.

METHODS: Eight medical students, with no prior laparoscopic skills, were recruited for a learning curve study where each student performed up to 10 FLS pattern cutting trials each day for 12 consecutive days. Following a two week break period, the students performed three pattern cutting trials to measure skill retention. Six medical students served as the control group and only performed pattern cutting trials on the first and last day. Cortical brain activation was measured in real time using fNIRS specific to the PFC, M1, and SMA regions. Pattern cutting scores were hand-calculated according the official FLS scoring metric and fNIRS data was analyzed through a set of robust and published post-processing suite called Homer2.

RESULTS: Figure1 indicates that the training FLS and control groups had a mean score of 29 ± 39 and 51.4 ± 36 for the first day, respectively. On the final retention test, the trained FLS group outperformed the control group (p=0.00) with scores of 223.5 ± 18 and 54.3 ± 45, respectively. Figure 2 shows that cotrical activation, as measured by the change in oxy-hemoglobin concentration in the cortex, decreases with respect to increasing days and is significantly lower than the control group (p<0.05). Furthermore, the trained FLS group shows a decrease in right lateral PFC activation and an increase in medial M1 activation compared to the control group (p<0.05, figures not shown).

Figure 1: FLS performance score learning curves

Figure 2: fNIRS activation learning curve

CONCLUSION: This is the first study to show motor skill learning based cortical changes via fNIRS for binaural dexterity laparoscopic skills. By showing that motor skill levels can be objectively determined by real time non-invasive cortical imaging of the PFC and M1, it is possible to compound traditional performance metric based evaluation to better determine the motor skill proficiency of a physician.

Presented at the SAGES 2017 Annual Meeting in Houston, TX.

Abstract ID: 79161

Program Number: S118

Presentation Session: MIS – Cool Stuff

Presentation Type: Podium