Intraoperative Ultrasound-Based Displacement Mapping Through Deep Learning
Whereas most previous studies have focused on MRI-to-iUS or MRI-to-MRI deformation modeling, our study demonstrates the feasibility of estimating spatially resolved brain shift directly from iUS-to-iUS scans using deep learning. This approach provides dense, real-time deformation fields for better intraoperative adaptability. Future work should expand on data set diversity and size, and integrate multitask learning to distinguish deformation from parenchymal collapse.
Authors
- Daniel Aaronson (ORCID: https://orcid.org/0000-0002-7197-6639)
- Vratko Himič (ORCID: https://orcid.org/0000-0002-3906-5317)
- Shrinit Babel (ORCID: https://orcid.org/0000-0002-6180-7778)
- Ajay M. Shah (ORCID: https://orcid.org/0000-0002-6547-0631)
- Ricardo J. Komotar
- Michael E. Ivan
- Zachary C. Gersey
Institutions
- Loma Linda University Medical Center (US)
- Johns Hopkins University (US)
- University of Miami (US)
- University of South Florida (US)
- Loma Linda University (US)
- Johns Hopkins Medicine (US)
- Neurological Surgery (US)
- Saint Louis University (US)
Publication Details
- Journal
- Operative Neurosurgery
- Published
- 2026-02-18
- DOI
- https://doi.org/10.1227/ons.0000000000001937
- Primary Topic
- Glioma Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 0.00