We evaluated whether, compared with conventional deep learning reconstruction (DLR) and zero-filling interpolation (ZIP), super-resolution DLR (SR-DLR) enhances the visualization of vestibular schwannomas and cranial nerves in three-dimensional constructive interference in steady state (3D-CISS) magnetic resonance (MR) imaging....
Super-resolution deep learning reconstruction shows promise for improving vestibular schwannoma visibility on 3D CISS MRI, but clinical adoption should await broader validation and regulatory clearance before replacing standard protocols.
Better MRI visualization of vestibular schwannomas could lead to earlier and more accurate diagnosis, influencing treatment planning decisions for patients with unexplained hearing loss or balance problems.
- 01Super-resolution deep learning reconstruction evaluated on 3D CISS MRI for vestibular schwannoma imaging.
- 02Published in Radiological Physics and Technology (DOI: 10.1007/s12194-026-01073-7).
- 03Compared AI-enhanced imaging to conventional MRI reconstruction methods.
- 04Goal is enhanced tumor visualization, which may aid diagnosis of inner ear tumors.
- 05Findings are preliminary and require external validation before clinical deployment.
Super-resolution deep learning reconstruction improves visualization of vestibular schwannomas on 3D CISS MRI compared to conventional methods.
studypartially supported- PMID
- 42223818
- DOI
- 10.1007/s12194-026-01073-7.
- Journal
- Radiological Physics and Technology
- Publication type
- research_article
- Evidence level
- 2b
- Population
- Patients with known or suspected vestibular schwannomas undergoing 3D CISS MRI
- Intervention
- Super-resolution deep learning reconstruction applied to 3D CISS MRI
- Comparator
- Conventional 3D CISS MRI reconstruction
Primary outcomes
Image quality and visualization of vestibular schwannomas on MRI