Normative hippocampal subfield volumes built from 34,789 MRI scans and applied to 3,521 patients
Authors built age-based reference volumes of the hippocampus and its subfields and used them to score individual patients with five neurological disorders.

A group of researchers from Chinese hospitals and universities, including Beijing Tiantan Hospital of Capital Medical University, has published normative reference values for the volumes of the hippocampus and its subfields in Nature Communications. The references are based on 3D T1-weighted magnetic resonance imaging (MRI) scans of 34,789 healthy people aged 4 to 85 years doi.org
Nature Communicationsdoi.orgOpen source ↗. The authors then used them to score 3,521 patients with neurological disorders
doi.org
Nature Communicationsdoi.orgOpen source ↗.
Normative references for hippocampal subfields
The hippocampus is a hub of cognition, and its subfields differ in how susceptible they are to aging and to neurological disease doi.org
Nature Communicationsdoi.orgOpen source ↗. The reference curves show that volume peaks at different ages in different subfields, from 27.4 to 40.1 years
doi.org
Nature Communicationsdoi.orgOpen source ↗.
The healthy-participant data came from the CHART consortium and from public neuroimaging collections, among them IXI, CamCAN, SALD, SRPBS, ABIDE, ADHD-200, NKI-RS, OASIS and curated OpenNeuro cohorts. Hippocampal subfield segmentation was done with FreeSurfer.
Centile scores in five disorders
For each patient, the team calculated a personalized deviation (centile) score relative to the reference. The patients had mild cognitive impairment, Alzheimer's disease, Parkinson's disease, cerebral small vessel disease or multiple sclerosis doi.org
Nature Communicationsdoi.orgOpen source ↗. According to the authors, the scores captured disease-specific hippocampal abnormalities and supported classification of the disorders
doi.org
Nature Communicationsdoi.orgOpen source ↗.
The scores were also associated with cognitive performance, with longitudinal cognitive outcomes and with disease progression across the disorders doi.org
Nature Communicationsdoi.orgOpen source ↗. External validation used data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and the Parkinson's Progression Markers Initiative (PPMI).
The authors also developed a personalized hippocampal health report intended to ease clinical translation doi.org
Nature Communicationsdoi.orgOpen source ↗. They conclude that hippocampal and subfield normative references can be used for individual diagnosis, cognitive assessment and prognostic evaluation in neurological diseases
doi.org
Nature Communicationsdoi.orgOpen source ↗. The material describes an analysis of imaging data; it does not report clinical use of the report.
Sources · 1
Found a mistake?Read also



