Abstract
Pedicle screw placement requires accurate reasoning about complex three-dimensional anatomy. While virtual reality has become a promising platform for surgical planning and guidance, existing systems typically communicate only a single recommended trajectory, providing limited information about the confidence associated with that recommendation. As artificial intelligence (AI) becomes increasingly integrated into surgical navigation, communicating confidence may improve human–AI
collaboration. We compare conventional deterministic guidance with two uncertainty-inspired guidance corridors that communicate confidence using blur and colour. As a proof of concept, we conducted a user study to evaluate performance, user perception, and visualization preference using parameterized rather than model-derived confidence. No significant differences were observed in entry-point error, tip-point error, task completion time, or screw adjustments between guidance representations.
Angular deviation differed significantly, with deterministic guidance producing the smallest deviation. Subjective trajectory-understanding ratings showed the largest numerical difference across conditions, although this difference was not statistically significant. Nevertheless, 11 of 14 participants preferred a confidence-aware representation. These findings support further investigation of confidence-aware visualization for future AI-assisted surgical navigation systems.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/COLAS_028.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/forum?id=XuepM8F24N
BibTex
@InProceedings{TouHam_ConfidenceAware_MICCAISAT2026,
author = { Toudeshky, Hamideh Hosseini AND Batmaz, Anil Ufuk AND Kersten-Oertel, Marta},
title = { { Confidence-Aware Visualization for Pedicle Screw Guidance in Virtual Reality } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS pending},
month = {pending},
page = {pending}
}