Deep learning networks reflect cytoarchitectonic features used in brain mapping

Abstract

The distribution of neurons in the cortex (cytoarchitecture) differs between cortical areas and constitutes the basis for structural maps of the human brain. Deep learning approaches provide a promising alternative to overcome throughput limitations of currently used cytoarchitectonic mapping methods, but typically lack insight as to what extent they follow cytoarchitectonic principles. We therefore investigated in how far the internal structure of deep convolutional neural networks trained for cytoarchitectonic brain mapping reflect traditional cytoarchitectonic features, and compared them to features of the current grey level index (GLI) profile approach. The networks consisted of a 10-block deep convolutional architecture trained to segment the primary and secondary visual cortex. Filter activations of the networks served to analyse resemblances to traditional cytoarchitectonic features and comparisons to the GLI profile approach. Our analysis revealed resemblances to cellular, laminar- as well as cortical area related cytoarchitectonic features. The networks learned filter activations that reflect the distinct cytoarchitecture of the segmented cortical areas with special regard to their laminar organization and compared well to statistical criteria of the GLI profile approach. These results confirm an incorporation of relevant cytoarchitectonic features in the deep convolutional neural networks and mark them as a valid support for high-throughput cytoarchitectonic mapping workflows.

AnNoBrainer, an Automated Annotation of Mouse Brain Images using Deep Learning

“…Important to note that to date Nissl, H&E and some IHC stains (e.g. α-synuclein) were successfully tested, yet AnNoBrainer was designed to also support multiplexed stains and can incorporate other IHC stains too, therefore can support a broader range of studies compared to previously published methods [11, 10] that focus primarily on traditional histology stains. L…”

Cytoarchitectonic Maps of the Human Metathalamus in 3D Space

“…To map the three subdivisions on every section, a deep-learning based brain mapping tool designed to map cytoarchitectonic structures in full stacks ( Schiffer et al, 2021c ) was applied. The deep-learning network architecture has shown to resemble cytoarchitectonic criteria ( Kiwitz et al, 2020 ) and has successfully been used to generate whole-stack maps of several cytoarchitectonic areas ( Schiffer et al, 2021c ). The method was trained on 57 delineated sections containing the MGB and its subdivisions.…”

A Review of Convolutional Neural Network Development in Computer Vision

“…At the same time, the function of neurons in the hidden layer is set to 0, which brings sparseness and makes it easy for the network to obtain sparse representation, reduce the number of parameters [50], and reduce overfitting. [51,52] Experiments show that ReLU has better performance than Sigmoid, and can be better to solve the gradient vanishing problem. The function formula of the ReLU is(6).…”