Project Summary
The hippocampus and surrounding medial temporal lobe (MTL) cortices are composed of several distinct
subregions that show differential vulnerability in aging and disease. In vivo measurement of these regions via
magnetic resonance imaging (MRI) is required for modern clinical research, and improving their measurement
may allow early detection and sensitive tracking of incipient pathology as well as facilitate differential diagnosis
of uncertain clinical presentation. Due to a lack of precise and sensitive measurement, the manner in which
MTL subregions change in aging, Alzheimer's disease (AD), and other neurological and psychiatric disorders is
uncertain. Discrepancies in how MTL subregions are segmented by different research groups further impedes
clinical translation of key research findings. The solution to this issue is to develop a reliable, validated,
harmonized protocol for segmentation of MTL subregions that can be applied to populations of variable age
and health, an initiative that was launched in 2013 by the Hippocampal Subfields Group (HSG). The HSG is
the only working group of its kind and includes over 200 scientists actively segmenting the human MTL. Our
primary deliverable will be a gold-standard, valid, and highly reliable MRI segmentation protocol that can be
broadly adopted and applied to various populations, enabling direct comparison of results across researchers.
Formal development of the protocol first began in 2015, before dedicated funding was available, when we
developed a segmentation protocol for subfields within the hippocampal body. In 2016-17, with funding from
the EU Joint Programme on Neurodegenerative Disease, we held three meetings during which we developed a
protocol for the hippocampal head. We are currently developing a protocol for the hippocampal tail, an effort
begun during a meeting in 2018 sponsored in part by the German Center for Neuro-degenerative Diseases.
We anticipate that the head, body, and tail protocol will be completed in 2023, after which we propose to host
a two-part meeting that will enable us to expand our protocol to include MTL cortical subregions
(entorhinal, perirhinal, and parahippocampal cortices). Given the importance of creating a harmonized
protocol that is both reliable and valid, this protocol must reflect the known anatomy of the human MTL. The
first half of the proposed meeting will therefore involve a series of lectures regarding MTL neuroanatomy and
histology, as well as hands-on dissection activities using human MTL specimens. This unique opportunity will
facilitate the second portion of the meeting, in which we propose to utilize labeled histology from HSG
neuroanatomists and the hands-on knowledge gained from the first portion of the meeting to (a) develop a draft
segmentation protocol for MRI, and (b) conduct an initial feasibility test by applying each segmentation rule to
publicly available MRI data sets collected from multiple ages and pathologies. The completed harmonized
protocol will reduce methodological discrepancies across research group and enhance in vivo clinical research
progress.