Skip to main content
Advertisement
  • Loading metrics

PLoS Computational Biology Issue Image | Vol. 20(1) February 2024

Extracting structure from feature values represented over point clouds

Point clouds are commonly used in many scientific fields. Each point of a cloud is typically associated with additional values or features. Thus, having the ability to identify if and how a given feature is structured along the cloud can provide insights for many applications. Here, we present the Structure Index, a graph-based method to quantify the distribution of features over data in arbitrary dimensional spaces. Spaces can be defined by neurons, time stamps, pixels, genes, etc... The image illustrates how the Structure Index eases the interpretation of neural representations from the brain head-direction system. Sebastian, Esparza and de la Prida 2024.

Image Credit: Julio Esparza

thumbnail
Extracting structure from feature values represented over point clouds

Point clouds are commonly used in many scientific fields. Each point of a cloud is typically associated with additional values or features. Thus, having the ability to identify if and how a given feature is structured along the cloud can provide insights for many applications. Here, we present the Structure Index, a graph-based method to quantify the distribution of features over data in arbitrary dimensional spaces. Spaces can be defined by neurons, time stamps, pixels, genes, etc... The image illustrates how the Structure Index eases the interpretation of neural representations from the brain head-direction system. Sebastian, Esparza and de la Prida 2024.

Image Credit: Julio Esparza

https://doi.org/10.1371/image.pcbi.v20.i01.g001