Editorial - Journal of Neuroinformatics and Neuroimaging (2021) Volume 6, Issue 3
Interneuron morphology representations for cell type discrimination: A systematic review.
In order to make individual morphologies accessible to normal statistics tools and machine learning algorithms, quantitative study of neuronal morphologies often begins with the selection of a specific feature representation. Many various feature representations, ranging from density maps to intersection profiles, have been proposed in the literature, but they have never been compared side by side. In this study, we conducted a thorough evaluation of different representations, assessing how well they captured the distinction between recognised morphological cell types. We employed multiple curated data sets of mouse retinal bipolar cells and cortical inhibitory neurons for our benchmarking attempt.Author(s): Shawn Kruger