2055 Toward Precision Education in Neurosurgical Training: Cognitive Load Estimation via Pupil Diameter
INTRODUCTION: The progression from resident to attending neurosurgeon requires developing technical and cognitive skills. Cognitive load (CL) can provide insight into the bandwidth required to perform tasks and whether residual capacity exists for increasing task burden. For neurosurgeons, the CL associated with routine tasks must be managed to allow decision-making under duress. The baseline change in pupil diameter (BCPD) can be used to calculate CL, providing insights into mental effort and decision-making processes. METHODS: We recorded residents and attendings performing craniotomies while wearing point-of-view and eye-tracking cameras. For the present analysis, we compared attending and resident CL during burr hole placement and bone flap elevation. We used the BCPD method, which computes a reference average pupil diameter established during scalp incision. CL is then calculated by measuring the average difference between this baseline and the current pupil diameter in the phase of interest. More negative numbers reflect lower CLs. RESULTS: During burr hole drilling, attending BCPD averaged -0.04 compared to -0.08 for residents. During bone flap elevation, attending BCPD averaged -0.27 versus -0.08 for residents. During sphenoid wing drilling, attending BCPD averaged -0.28 versus 0.2 for residents. During dural opening attending BCPD averaged -0.25 versus 0.03 for residents. CONCLUSIONS: Our findings suggest that attendings experience lower CL during more complex tasks. Incorporating measures of CL alongside other machine learning-enabled metrics can identify areas where trainees may need additional support. This integrated approach creates a new paradigm for precision education in neurosurgery. Next steps include integrating the remainder of our video dataset into the model, assessing CL throughout the remainder of the procedure, and incorporating gaze patterns into our models.
Authors
- Alice Hung (ORCID: https://orcid.org/0000-0001-6815-3765)
- Risheng Xu (ORCID: https://orcid.org/0000-0003-4288-1232)
- Christopher M. Jackson (ORCID: https://orcid.org/0000-0002-5958-2004)
- S. Swaroop Vedula (ORCID: https://orcid.org/0000-0002-6992-2957)
- Judy Huang (ORCID: https://orcid.org/0000-0002-0675-1935)
- Jose L. Porras (ORCID: https://orcid.org/0000-0001-8664-7646)
- Aidan Leong (ORCID: https://orcid.org/0000-0003-2853-404X)
- Roger D. Soberanis-Mukul (ORCID: https://orcid.org/0000-0003-0712-0534)
- Henry Brem (ORCID: https://orcid.org/0000-0001-7539-4786)
- Andrew S. Luksik
- Masaru Ishii (ORCID: https://orcid.org/0000-0002-4215-007X)
- Brendan F. Judy (ORCID: https://orcid.org/0000-0002-4981-1408)
- Jordina N. Rincon Torroella
- Chetan Bettegowda (ORCID: https://orcid.org/0000-0001-9991-7123)
- Nancy Abu-Bonsrah (ORCID: https://orcid.org/0000-0001-8195-1622)
- Mathias Unberath (ORCID: https://orcid.org/0000-0002-0055-9950)
- Gary L. Gallia (ORCID: https://orcid.org/0000-0001-6438-7329)
- Jiachen Liu (ORCID: https://orcid.org/0000-0002-0263-3704)
- Rohit Shankar
- Matthew J. Cullen
- K. Joshi
Publication Details
- Journal
- Neurosurgery
- Published
- 2025-03-14
- DOI
- https://doi.org/10.1227/neu.0000000000003360_2055
- Primary Topic
- Surgical Simulation and Training
- Type
- article
- Field-Weighted Citation Impact
- 0.00