Patient-Specific Virtual Reality for Lumbar Decompression Surgery: An Automated Imaging Pipeline and Educational Assessment

Manual construction of patient-specific models has limited the use of virtual reality (VR) for case-specific spine surgical rehearsal, and simulator assessments are often not patient-specific. We present a largely automated pipeline that converts routine preoperative CT and MRI into patient-specific 3D models for VR rehearsal of lumbar decompression, paired with case-specific written assessments. CT was segmented with an unmodified vertebral network and MRI with an in-house nnU-Net; volumes were registered and exported for VR. A locally hosted large language model generated per-case assessments from each patient's radiology report; the pipeline ran in approximately two minutes per case, and MRI segmentation reached a mean Dice of 0.870. Eight patient-specific models were prepared, and eleven trainees (undergraduate to fellow) completed a faculty-guided session with pre/post knowledge testing. Knowledge scores increased from 5.82 to 7.27 out of 10 (Wilcoxon $p=0.008$; Cohen's $d=1.01$), with 8 of 11 improving and none declining, and trainees rated educational value favorably. This feasibility study demonstrates a fast, fully private pipeline producing patient-specific VR models and matched assessments; the educational findings are preliminary and uncontrolled, motivating a larger controlled trial.

Authors

Institutions

Publication Details

Published
2026-11-17
DOI
https://doi.org/10.22541/authorea.15012338/v1
Primary Topic
Anatomy and Medical Technology
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Patient-Specific Virtual Reality for Lumbar Decompression Surgery: An Automated Imaging Pipeline and Educational Assessment

Tayler Ross, Raj Kumar Ranabhat, Michael Hardisty, Joel A. Finkelstein et al.
2 links
Anatomy and Medical Technology
preprint

Patient-Specific Virtual Reality for Lumbar Decompression Surgery: An Automated Imaging Pipeline and Educational Assessment

Tayler Ross, Raj Kumar Ranabhat, Michael Hardisty, Joel A. Finkelstein, Ran Ankory, Stanley Hamstra, Yu Yang Jiao
preprint en

Abstract

Manual construction of patient-specific models has limited the use of virtual reality (VR) for case-specific spine surgical rehearsal, and simulator assessments are often not patient-specific. We present a largely automated pipeline that converts routine preoperative CT and MRI into patient-specific 3D models for VR rehearsal of lumbar decompression, paired with case-specific written assessments. CT was segmented with an unmodified vertebral network and MRI with an in-house nnU-Net; volumes were registered and exported for VR. A locally hosted large language model generated per-case assessments from each patient's radiology report; the pipeline ran in approximately two minutes per case, and MRI segmentation reached a mean Dice of 0.870. Eight patient-specific models were prepared, and eleven trainees (undergraduate to fellow) completed a faculty-guided session with pre/post knowledge testing. Knowledge scores increased from 5.82 to 7.27 out of 10 (Wilcoxon $p=0.008$; Cohen's $d=1.01$), with 8 of 11 improving and none declining, and trainees rated educational value favorably. This feasibility study demonstrates a fast, fully private pipeline producing patient-specific VR models and matched assessments; the educational findings are preliminary and uncontrolled, motivating a larger controlled trial.

Sunnybrook Health Science Centre (CA), University of Toronto (CA), Sunnybrook Hospital (CA), Sunnybrook Research Institute (CA)
University of Toronto
Good health and well-being
Anatomy and Medical Technology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon