Loading…

Developing an anatomically valid segmentation protocol for early tau regions: the anterior medial temporal lobe cortices

Background The anterior portion of the medial temporal lobe (MTL) is one of the earliest regions showing tau deposition in Alzheimer’s disease (AD) and hence a key focus area for AD imaging biomarkers. However, the anatomical variability of the anterior MTL presents a challenge for segmentation prot...

Full description

Saved in:
Bibliographic Details
Published in:Alzheimer's & dementia 2023-12, Vol.19 (S17), p.n/a
Main Authors: Sadeghpour, Niyousha, Ittyerah, Ranjit, Denning, Amanda E, Trotman, Winifred, Lim, Sydney A, Chung, Eunice, Sadaghiani, Shokufeh, Prabhakaran, Karthik, Bedard, Madigan L, Ohm, Daniel T, Schuck, Theresa, Artacho‐Perula, Emilio, Martin, Maria Mercedes Iniguez de Onzono, Munoz, Monica, Romero, Francisco Javier Molina, Gonzalez, Jose Carlos Delgado, del Mar Arroyo Jimenez, Maria, del Pilar Marcos Rabal, Maria, Sanchez, Sandra Cebada, de la Rosa Prieto, Carlos, Parada, Marta Corcoles, Insausti, Ricardo, Wuestefeld, Anika, McMillan, Corey T, Grossman, Murray, Lee, Eddie B, Detre, John A., Trojanowski, John Q, Tisdall, Dylan M, Irwin, David J., Wolk, David A., Yushkevich, Paul A., Wisse, Laura EM
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Background The anterior portion of the medial temporal lobe (MTL) is one of the earliest regions showing tau deposition in Alzheimer’s disease (AD) and hence a key focus area for AD imaging biomarkers. However, the anatomical variability of the anterior MTL presents a challenge for segmentation protocols. Leveraging a novel postmortem dataset with histology and MRI, we aimed to develop a histologically‐informed segmentation protocol for the anterior entorhinal cortex (ERC), Brodmann Area (BA) 35 (approximates transentorhinal cortex), and BA36 for in vivo 3 tesla (T) MRI. Method Digitized 50‐µm thick MTL Nissl‐stained coronal histology sections from 20 cases (Table 1) were annotated by expert neuroanatomists. Cases with and without neurodegenerative diseases were included to ensure broad generalizability of the protocol. The histology sections were registered to same‐subject 0.2×0.2×0.2‐mm3 9.4 T postmortem MRI and were analyzed together to determine the location of the histological borders of the MTL cortices in relation to anatomical landmarks observable on MRI. Result The distance between the anterior histological border of ERC, BA35 and BA36 was assessed in relation to different anatomical landmarks observable on MRI. The distance relative to the hippocampus was chosen, as there was relatively low between‐subject variability in distance (low SD in Table 2) and this region is easily identifiable on MRI. The ERC starts 4.75 mm (median) anterior to the hippocampus, BA35 9.25 mm and BA36 10.25 mm. There was no significant difference in distances between patients with and without neurodegenerative disease. Next, we analyzed the location of the ERC on histology sections. We determined that regardless of the depth of the collateral sulcus (CS) (contrary to more posterior regions), the inferior border of the ERC is on average at the edge of the CS and the superior border at the superior edge of the parahippocampal gyrus (Figure 1a), but at the halfway point for the first section with ERC. These segmentation rules for ERC can be applied to T1‐MRI (Figure 1b). Conclusion Segmentation rules for anterior BA35 and BA36 are under development. This histologically‐informed segmentation protocol is expected to improve the efficacy of imaging biomarkers for tau deposition in AD.
ISSN:1552-5260
1552-5279
DOI:10.1002/alz.075831