Unsupervised machine learning identifies clinically relevant patterns of CSF dynamic dysfunction in normal pressure hydrocephalus
NMF-generated patterns of CSF distribution showed high accuracy in discerning individuals with iNPH from controls. The patterns most relying on DESH features showed highest potential for independently predicting NPH diagnosis. The algorithm we proposed should not be perceived as a replacement for human expertise but rather as an additional tool to assist clinicians in achieving accurate diagnoses.
Authors
- Benjamin D. Elder (ORCID: https://orcid.org/0000-0002-7782-7829)
- Clifford R. Jack (ORCID: https://orcid.org/0000-0001-7916-622X)
- Matthew C. Murphy
- Petrice M. Cogswell (ORCID: https://orcid.org/0000-0002-8128-4476)
- Emanuele Camerucci (ORCID: https://orcid.org/0000-0002-5510-9834)
- Jeffrey L. Gunter (ORCID: https://orcid.org/0000-0001-7813-9078)
- John Huston (ORCID: https://orcid.org/0000-0003-0584-9460)
- Jeremy K. Cutsforth‐Gregory (ORCID: https://orcid.org/0000-0002-5689-4424)
- Hugo Botha (ORCID: https://orcid.org/0000-0003-4390-685X)
- Matthew L. Senjem (ORCID: https://orcid.org/0000-0001-9308-9275)
- David Thomas Jones (ORCID: https://orcid.org/0000-0002-4807-9833)
- Ignacio Jusué-Torres (ORCID: https://orcid.org/0000-0002-9749-1912)
- Jonathan Graff-Radford
Institutions
- Mayo Clinic (US)
- Mayo Clinic Health System (US)
- Mayo Clinic in Arizona (US)
- Mayo Clinic in Florida (US)
Publication Details
- Journal
- Clinical Neurology and Neurosurgery
- Published
- 2025-09-16
- DOI
- https://doi.org/10.1016/j.clineuro.2025.109162
- Primary Topic
- Cerebrospinal fluid and hydrocephalus
- Type
- article
- Field-Weighted Citation Impact
- 0.00