Smooth Principal Component Analysis over two-dimensional manifolds with an application to Neuroimaging

@article{Lila2016SmoothPC,
  title={Smooth Principal Component Analysis over two-dimensional manifolds with an application to Neuroimaging},
  author={Eardi Lila and John A. D. Aston and Laura M. Sangalli},
  journal={arXiv: Applications},
  year={2016}
}
Motivated by the analysis of high-dimensional neuroimaging signals located over the cortical surface, we introduce a novel Principal Component Analysis technique that can handle functional data located over a two-dimensional manifold. For this purpose a regularization approach is adopted, introducing a smoothing penalty coherent with the geodesic distance over the manifold. The model introduced can be applied to any manifold topology, can naturally handle missing data and functional samples… 
Statistical Analysis of Functions on Surfaces, With an Application to Medical Imaging
  • E. Lila, J. Aston
  • Mathematics, Computer Science
    Journal of the American Statistical Association
  • 2019
TLDR
This work introduces a comprehensive framework for the analysis of functional data, whose domain is a two-dimensional manifold and the domain itself is subject to variability from sample to sample, and proposes an associated estimation framework.
Manifold Data Analysis with Applications to High-Frequency 3D Imaging
TLDR
A framework that converts manifolds into functional objects, an efficient 2-step functional principal component method, and a manifold-on-scalar regression model is proposed, used to understand how individual characteristics, such as age and genetic ancestry, influence the shape of the human face.
Principal component analysis for functional data on Riemannian manifolds and spheres
Functional data analysis on nonlinear manifolds has drawn recent interest. Sphere-valued functional data, which are encountered for example as movement trajectories on the surface of the earth, are
Functional and Geometric Statistical Analysis of Textured Surfaces with an application to Medical Imaging
TLDR
This work introduces a comprehensive framework for the analysis of functional data, whose domain is a two-dimensional manifold and the domain itself is subject to variability from sample to sample.
Interpretable discriminant analysis for functional data supported on random non-linear domains
TLDR
A novel framework for the classification of functional data supported on non-linear, and possibly random, manifold domains and is able to estimate discriminant directions that capture both cortical geometric and thickness predictive features of Alzheimer’s Disease, which are consistent with the existing neuroscience literature.
Statistics on functional data and covariance operators in linear inverse problems
TLDR
The proposed model is applied to MEG data, leading to a novel statistical approach to the investigation of functional connectivity, and exploiting the spatial structure of the sample data by introducing a flexible regularizing term embedded in the model.
Modelling spatially dependent functional data via regression with di ff erential regularization
We propose a method for modelling spatially dependent functional data, based on regression with differential regularization. The regularizing term enables to include problem-specific information
Exploratory Analysis of Brain Signals through Low Dimensional Embedding
TLDR
The approach is to extract the optimal lower dimensional representations for each brain region and then characterize and estimate connectivity between regions through these factors to investigate consistency of these factors across all epochs during the entire resting-state period.
Functional Regression with Unknown Manifold Structures
Statistical methods that adapt to unknown population structures are attractive due to both practical and theoretical advantages over their non-adaptive counterparts. We contribute to adaptive
...
...

References

SHOWING 1-10 OF 52 REFERENCES
Principal Component Analysis of Two-Dimensional Functional Data
This article presents and compares two approaches of principal component (PC) analysis for two-dimensional functional data on a possibly irregular domain. The first approach applies the singular
Statistical Analysis on Brain Surfaces
TLDR
This case study will determine if there is an effect of family income on the growth of hippocampus in children in detail and the methods described are illustrated with the hippocampus surface data set published in Chung et al. 2011.
Large-scale Probabilistic Functional Modes from resting state fMRI
Spatial regression models over two-dimensional manifolds
We propose a regression model for data spatially distributed over general two-dimensional Riemannian manifolds. This is a generalized additive model with a roughness penalty term involving a
Smoothed functional principal components analysis by choice of norm
The principal components analysis of functional data is often enhanced by the use of smoothing. It is shown that an attractive method of incorporating smoothing is to replace the usual L 2
Functional principal components analysis via penalized rank one approximation
TLDR
An alternative approach to FPCA using penalized rank one approximation to the data matrix is proposed, which naturally incorporates spline smoothing of discretized functional data and suggests an efficient power algorithm for computation.
Graph Regularized Nonnegative Matrix Factorization for Data Representation.
TLDR
In GNMF, an affinity graph is constructed to encode the geometrical information and a matrix factorization is sought, which respects the graph structure, and the empirical study shows encouraging results of the proposed algorithm in comparison to the state-of-the-art algorithms on real-world problems.
Graph Regularized Nonnegative Matrix Factorization for Data Representation
...
...