% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@INPROCEEDINGS{Bludau:1048761,
author = {Bludau, Sebastian and Dickscheid, Timo and Schiffer,
Christian and Steffens, Anna and Upschulte, Eric and Amunts,
Katrin},
title = {{L}ayer-specific cell counts in {B}ig{B}rain –
decomposing cortex-wide numbers based on cytoarchitectonics},
reportid = {FZJ-2025-04876},
year = {2025},
abstract = {<b>Introduction:</b> Cell counts of the cerebral cortex
represent one of the most fundamental characteristics of
brain organization, and serve as the basis for studying
evolution, development and disease (e.g., [1, 2]). However,
the total number of cells in the brain or cerebral cortex
does not reflect the variation between layers and cortical
areas, which is related to the functional heterogeneity of
the human brain. Therefore, we investigated layer- specific
cell counts in 94 cytoarchitectonic cortical areas in the
anatomical BigBrain model [3].<br><br><b>Methods:</b> The
study is based on cell counts and cortical thickness
measurements in high- resolution 2D scans (1 μm in-plane
resolution), most of which are already publicly available
[4]. In total, 940 patches were analyzed across 94 areas of
the Julich Brain Atlas [5] (Fig. 1), with 10 patches sampled
per area, ensuring a $>65\%$ probability of region being
present at each location. Using siibra-python [6], patches
were assigned to Julich Brain cytoarchitectonic
probabilistic maps by transforming their locations to MNI
space [7]. These were chosen to be perpendicular to the
cortical surface. Manual layer annotations were performed by
anatomical experts and independently verified. Cells were
segmented using a novel deep learning approach [8, 9]
including correction for truncated cells (Fig1). Cell
numbers were corrected for histological shrinkage [10]. To
cross-validate data and capture intersubject variability
among brains, patches from frontal pole area Fp1, motor area
4a, and visual area hOc1of ten Julich- atlas brains were
analyzed using the same counting approach and compared to
the BigBrain data. <br><br><b>Results:</b> Comprehensive
data sets of 940 image patches were obtained including the 1
μm images with manual annotation of cortical layers, the
cell counts, cell sizes, and cortical as well as layer
thicknesses (Fig1, Fig 2). The analysis revealed a
considerable regional variations in cell counts across
layers and areas as illustrated in Fig. 2. E.g., areas of
the insula showed up to $15\%$ variance. After shrinkage
correction and adjusting for truncated cells, the corrected
total cell count of the human cortex has been estimated to
be approximately 33.9 billion. Based on a neuron-to-glia
ratio of 1:1.5 [1], this corresponds to 13.6 billion neurons
and 20.3 billion glial cells. The average cortical thickness
across all regions was 2667.15 μm. Quantitative measures of
areas Fp1 of the BigBrain (Fig. 2) and of the other two
areas were in the range of variation of the ten brains from
Julich Brain Atlas. The data will be shared as part of a
growing dataset collection [11] in accordance with the FAIR
principles via the EBRAINS
infrastructure.<br><br><b>Discussion:</b> This study has
introduced a new, comprehensive dataset with detailed area-
and layer specific cell counts of the human cerebral cortex,
supplementing previous data at whole- cortex level [1]. It
extended our knowledge on cytoarchitectonic differences,
e.g., etween granular, dysgranular and granular areas and
further quantified regional differences at laminar level.
The considerable differences between areas within the
insular cortex, as one example, confirms the hypothesis that
macroanatomically defined regions do not adequately reflect
the microstructural organization of the brain; they may lump
together structurally and functionally different areas. Data
on cortical thickness correspond to earlier histological and
MR-based findings [2, 12, 13]. We will continuously
supplement the data together with new releases of Julich
Brain Atlas areas. Cell counts based on cytoarchitectonics
may serve as reference for comparative and disease studies,
and inform modeling and simulation, and AI, highlighting the
value of high-resolution atlases for capturing details of
microscopical brain
organization.<br><br><b>References:</b><ol><li>von Bartheld,
C.S., et al., The search for true numbers of neurons and
glial cells in the human brain: A review of 150 years of
cell counting. J Comp Neurol, 2016. 524(18): p.
3865-3895.<li>von Economo, C.F., et al., Die
Cytoarchitektonik der Hirnrinde des erwachsenen Menschen.
1925: J. Springer.<li>Amunts, K., et al., BigBrain: an
ultrahigh-resolution 3D human brain model. Science,
340(6139): p. 1472-5.<li>Schiffer, C., Lepage, C.,
Omidyeganeh, M., Mohlberg, H., Brandstetter, A., Bludau, S.,
Heuer, K., Toussaint, P.-J., Wenzel, S., Dickscheid, T.,
Evans, A. C., $\&$ Amunts, K. , Selected 1 micron scans of
BigBrain histological sections (v1.0) 2022. </li><li>Amunts,
K., et al., Julich-Brain: A 3D probabilistic atlas of the
human brain's cytoarchitecture. Science, 2020. 369(6506): p.
988-992. </li><li>Dickscheid, T., Gui, X., Simsek, A. N.,
Koehnen, L., Marcenko, V., Schiffer, C., Bludau, S., $\&$
Amunts, K., siibra-python (Zenodo).
https://zenodo.org/records/14184565. </li><li>Lebenberg, J.,
et al., A framework based on sulcal constraints to align
preterm, infant and adult human brain images acquired in
vivo and post mortem. Brain Struct Funct, 223(9): p.
4153-4168. </li><li>Ma, J., et al., The multimodality cell
segmentation challenge: toward universal solutions. Nat
Methods, 2024. 21(6): p. 1103-1113. </li><li>Upschulte, E.,
et al., Contour proposal networks for biomedical instance
segmentation. Med Image Anal, 2022. 77: p. 102371.
</li><li>Amunts, K., et al., Gender-specific left-right
asymmetries in human visual cortex. J Neurosci, 2007. 27(6):
p. 1356-64. </li><li>Dickscheid, T., Bludau, S., Paquola,
C., Schiffer, C., Upschulte, E., $\&$ Amunts, K.,
Layerspecific distributions of segmented cells in different
cytoarchitectonic regions of BigBrain iso cortex. EBRAINS
project:
https://search.kg.ebrains.eu/instances/f06a2fd1-a9ca-42a3-b754-adaa025adb10.
</li><li>Frangou, S., et al., Cortical thickness across the
lifespan: Data from 17,075 healthy individuals aged 3-90
years. Hum Brain Mapp, 2022. 43(1): p. 431-451.
</li><li>Wagstyl, K., et al., BigBrain 3D atlas of cortical
layers: Cortical and laminar thickness gradients diverge in
sensory and motor cortices. PLoS Biol, 2020. 18(4): p.
e3000678. </li></ol>},
month = {Dec},
date = {2025-12-08},
organization = {EBRAINS summit 2025, Brüssel
(Belgium), 8 Dec 2025 - 11 Dec 2025},
subtyp = {After Call},
cin = {INM-1},
cid = {I:(DE-Juel1)INM-1-20090406},
pnm = {5241 - Molecular Information Processing in Cellular Systems
(POF4-524) / 5254 - Neuroscientific Data Analytics and AI
(POF4-525) / EBRAINS 2.0 - EBRAINS 2.0: A Research
Infrastructure to Advance Neuroscience and Brain Health
(101147319) / HIBALL - Helmholtz International BigBrain
Analytics and Learning Laboratory (HIBALL) (InterLabs-0015)},
pid = {G:(DE-HGF)POF4-5241 / G:(DE-HGF)POF4-5254 /
G:(EU-Grant)101147319 / G:(DE-HGF)InterLabs-0015},
typ = {PUB:(DE-HGF)24},
url = {https://juser.fz-juelich.de/record/1048761},
}