Automated Midline Shift Detection in Head CT Using Localization and Symmetry Techniques Based on User-Selected Slice
We developed an AI model that reliably identifies the lateral ventricles and the cerebral midline across various pathologies in patient CT scans. Most importantly, our model accurately identifies and stratifies clinically significant and emergent MLS from nonemergent cases. This could serve as a foundational element for a future clinically integrated approach that flags urgent studies for expedited review, potentially facilitating more timely treatment when necessary.
Authors
- Hrithwik Shalu (ORCID: https://orcid.org/0000-0002-8944-434X)
- Vaios Hatzoglou (ORCID: https://orcid.org/0000-0003-0252-0694)
- Nathaniel Swinburne (ORCID: https://orcid.org/0000-0001-5831-7287)
- N. Banayan
- Andrei I. Holodny (ORCID: https://orcid.org/0000-0002-1159-2705)
- Joseph N. Stember (ORCID: https://orcid.org/0000-0003-3169-9590)
- Zhigang Zhang (ORCID: https://orcid.org/0000-0002-1556-7945)
Institutions
- Memorial Sloan Kettering Cancer Center (US)
- Indian Institute of Technology Madras (IN)
Publication Details
- Journal
- American Journal of Neuroradiology
- Published
- 2025-08-20
- DOI
- https://doi.org/10.3174/ajnr.a8767
- Citations
- 1
- Primary Topic
- Traumatic Brain Injury and Neurovascular Disturbances
- Type
- article
- Field-Weighted Citation Impact
- 1.39