Deep Learning-Based Classification of GAD67-Positive Neurons Without the Immunosignal

Abstract

Excitatory neurons and GABAergic interneurons constitute neural circuits and play important roles in information processing. In certain brain regions, such as the neocortex and the hippocampus, there are fewer interneurons than excitatory neurons. Interneurons have been quantified via immunohistochemistry, for example, for GAD67, an isoform of glutamic acid decarboxylase. Additionally, the expression level of other proteins varies among cell types. For example, NeuN, a commonly used marker protein for postmitotic neurons, is expressed differently across brain regions and cell classes. Thus, we asked whether GAD67-immunopositive neurons can be detected using the immunofluorescence signals of NeuN and the fluorescence signals of Nissl substances. To address this question, we stained neurons in layers 2/3 of the primary somatosensory cortex (S1) and the primary motor cortex (M1) of mice and manually labeled the neurons as either cell type using GAD67 immunosignals. We then sought to detect GAD67-positive neurons without GAD67 immunosignals using a custom-made deep learning-based algorithm. Using this deep learning-based model, we succeeded in the binary classification of the neurons using Nissl and NeuN signals without referring to the GAD67 signals. Furthermore, we confirmed that our deep learning-based method surpassed classic machine-learning methods in terms of binary classification performance. Combined with the visualization of the hidden layer of our deep learning algorithm, our model provides a new platform for identifying unbiased criteria for cell-type classification.

Machine learning-based segmentation of the rodent hippocampal CA2 area from Nissl-stained sections

“…Before images of each set of slices were captured, the laser power for each fluorescence channel was set just below the intensity that would cause fluorescence saturation ( Yamashiro et al, 2021 ). The images (1024 × 1024 pixels, 16-bit intensity) were acquired at a Z-interval of 1.0 μm using an FV1200 confocal microscope (Olympus, Tokyo, Japan) equipped with 20× objectives and Z-stacked using ImageJ software (National Institutes of Health, MD, USA).…”

Rodent control of chamber brightness leads to the emergence of unique circadian rhythms

“…The images (1,024 × 1,024 pixels, 16-bit intensity) of each region of interest (ROI) were acquired at Zintervals of 2.0 µm (at 10× or 40× magni cation) using a confocal microscope (FV1000, Olympus, Tokyo, Japan) equipped with 10× and 40× objective lenses [43][44][45][46] ; the 10× lens was mainly used, but in some cases, a 40× lens was also used for scrupulous visual inspection. The images were then Z-stacked using ImageJ software (National Institutes of Health, MD, USA).…”

Deep-learning based breast cancer detection for cross-staining histopathology images

“…However, in clinical research, a typical study cohort for target protein expression ranges from tens to hundreds; such a small number of datasets is insufficient for high performance model development, and the shortage of public datasets provides limited support in transfer learning. The data requirement sets a high entry barrier for adopting AI-DL technology in clinical research with special stains such as fluorescent stain [ 19 , 20 ]. One conceivable approach for filling this gap is to use color normalization algorithms to compensate for model variation in different staining methodologies, but previous studies have mostly focused on solving the variation in the same colored detection methods [ [21] , [22] , [23] ].…”