1. Construction of embedded fMRI resting-state functional connectivity networks using manifold learning
- Author
-
Constantinos I. Siettos, Ioannis Gallos, Evangelos Galaris, Gallos, I. K., Galaris, E., and Siettos, Konstantinos
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Computer science ,Cognitive Neuroscience ,Diffusion map ,Functional connectivity networks ,Kernel principal component analysis ,Machine Learning (cs.LG) ,03 medical and health sciences ,0302 clinical medicine ,Machine learning ,FOS: Mathematics ,Mathematics - Numerical Analysis ,Multidimensional scaling ,Resting-state fMRI ,030304 developmental biology ,Functional connectivity network ,0303 health sciences ,Numerical Analysis ,Resting state fMRI ,business.industry ,Nonlinear dimensionality reduction ,Pattern recognition ,Numerical Analysis (math.NA) ,Euclidean distance ,Manifold learning ,Quantitative Biology - Neurons and Cognition ,FOS: Biological sciences ,Metric (mathematics) ,Benchmark (computing) ,Schizophrenia ,Neurons and Cognition (q-bio.NC) ,Artificial intelligence ,business ,030217 neurology & neurosurgery ,Research Article - Abstract
We construct embedded functional connectivity networks (FCN) from benchmark resting-state functional magnetic resonance imaging (rsfMRI) data acquired from patients with schizophrenia and healthy controls based on linear and nonlinear manifold learning algorithms, namely, Multidimensional Scaling, Isometric Feature Mapping, Diffusion Maps, Locally Linear Embedding and kernel PCA. Furthermore, based on key global graph-theoretic properties of the embedded FCN, we compare their classification potential using machine learning. We also assess the performance of two metrics that are widely used for the construction of FCN from fMRI, namely the Euclidean distance and the cross correlation metric. We show that diffusion maps with the cross correlation metric outperform the other combinations.
- Published
- 2021