Rogers F Silva

University of New Mexico, Electrical and Computer Engineering, Graduate Student

Followers

Following

Co-authors

Public Views

I am is a multidisciplinary research scientist developing algorithms for statistical and machine learning, image analysis, numerical optimization, memory efficient large scale data reduction, and distributed analyses, mostly working with multimodal, multi-subject neuroimaging data from thousands of subjects.
I received a B.Sc. degree in electrical engineering in 2003 from the Catholic University (PUCRS), Porto Alegre, RS, Brazil, a M.S. degree in computer engineering (with minors in statistics and in mathematics) in 2011, and a Ph.D. degree in computer engineering in 2016 (expected), both from The University of New Mexico, Albuquerque, NM, USA.
I have worked as an engineer, lecturer, and consultant before initiating my graduate studies, and I am currently a senior graduate research assistant at The Mind Research Network, Albuquerque, NM, USA.
My interests are: multimodal data fusion, statistical and machine learning, image, video and data analysis, multiobjective, combinatorial and constrained optimization, signal processing, and neuroimaging.
Supervisors: Vince D. Calhoun, Marios S. Pattichis, and Sergey M. Plis

less

InterestsView All (8)

Uploads

Papers by Rogers F Silva

A baseline for the multivariate comparison of resting-state networks

by Erik B Erhardt, Eswar Damaraju, and Rogers F Silva

Frontiers in Systems Neuroscience

As the size of functional and structural MRI datasets expands, it becomes increasingly important ... more As the size of functional and structural MRI datasets expands, it becomes increasingly important to establish a baseline from which diagnostic relevance may be determined, a processing strategy that efficiently prepares data for analysis, and a statistical approach that identifies important effects in a manner that is both robust and reproducible. In this paper, we introduce a multivariate analytic approach that optimizes sensitivity and reduces unnecessary testing. We demonstrate the utility of this mega-analytic approach by identifying the effects of age and gender on the resting-state networks (RSNs) of 603 healthy adolescents and adults (mean age: 23.4 years, range: 12-71 years). Data were collected on the same scanner, preprocessed using an automated analysis pipeline based in SPM, and studied using group independent component analysis. RSNs were identified and evaluated in terms of three primary outcome measures: time course spectral power, spatial map intensity, and functiona...

Download

Spatiospectral Decomposition of Multi-subject EEG: Evaluating Blind Source Separation Algorithms on Real and Realistic Simulated Data

by Rogers F Silva and David Bridwell

Brain topography, Jan 24, 2016

Electroencephalographic (EEG) oscillations predominantly appear with periods between 1 s (1 Hz) a... more Electroencephalographic (EEG) oscillations predominantly appear with periods between 1 s (1 Hz) and 20 ms (50 Hz), and are subdivided into distinct frequency bands which appear to correspond to distinct cognitive processes. A variety of blind source separation (BSS) approaches have been developed and implemented within the past few decades, providing an improved isolation of these distinct processes. Within the present study, we demonstrate the feasibility of multi-subject BSS for deriving distinct EEG spatiospectral maps. Multi-subject spatiospectral EEG decompositions were implemented using the EEGIFT toolbox ( http://mialab.mrn.org/software/eegift/ ) with real and realistic simulated datasets (the simulation code is available at http://mialab.mrn.org/software/simeeg ). Twelve different decomposition algorithms were evaluated. Within the simulated data, WASOBI and COMBI appeared to be the best performing algorithms, as they decomposed the four sources across a range of component n...

Memory Efficient PCA Methods for Large Group ICA

Frontiers in Neuroscience, 2016

Principal component analysis (PCA) is widely used for data reduction in group independent compone... more Principal component analysis (PCA) is widely used for data reduction in group independent component analysis (ICA) of fMRI data. Commonly, group-level PCA of temporally concatenated datasets is computed prior to ICA of the group principal components. This work focuses on reducing very high dimensional temporally concatenated datasets into its group PCA space. Existing randomized PCA methods can determine the PCA subspace with minimal memory requirements and, thus, are ideal for solving large PCA problems. Since the number of dataloads is not typically optimized, we extend one of these methods to compute PCA of very large datasets with a minimal number of dataloads. This method is coined multi power iteration (MPOWIT). The key idea behind MPOWIT is to estimate a subspace larger than the desired one, while checking for convergence of only the smaller subset of interest. The number of iterations is reduced considerably (as well as the number of dataloads), accelerating convergence without loss of accuracy. More importantly, in the proposed implementation of MPOWIT, the memory required for successful recovery of the group principal components becomes independent of the number of subjects analyzed. Highly efficient subsampled eigenvalue decomposition techniques are also introduced, furnishing excellent PCA subspace approximations that can be used for intelligent initialization of randomized methods such as MPOWIT. Together, these developments enable efficient estimation of accurate principal components, as we illustrate by solving a 1600-subject group-level PCA of fMRI with standard acquisition parameters, on a regular desktop computer with only 4 GB RAM, in just a few hours. MPOWIT is also highly scalable and could realistically solve group-level PCA of fMRI on thousands of subjects, or more, using standard hardware, limited only by time, not memory. Also, the MPOWIT algorithm is highly parallelizable, which would enable fast, distributed implementations ideal for big data analysis. Implications to other methods such as expectation maximization PCA (EM PCA) are also presented. Based on our results, general recommendations for efficient application of PCA methods are given according to problem size and available computational resources. MPOWIT and all other methods discussed here are implemented and readily available in the open source GIFT software.

Large scale collaboration with autonomy: Decentralized data ICA

by Rogers F Silva and Sergey Plis

2015 IEEE 25th International Workshop on Machine Learning for Signal Processing (MLSP), 2015

Comparison of PCA approaches for group ICA

NeuroImage, 2015

 Group ICA of fMRI on very large data sets is becoming more common  GIFT (since 2009) and MELOD... more  Group ICA of fMRI on very large data sets is becoming more common  GIFT (since 2009) and MELODIC (since 2014) enable analysis of thousands of subjects  We compare ten available approaches including a Pareto-optimal analysis  We provide new analyses and comments on "Group-PCA for very large fMRI datasets"

Download

The tenth annual MLSP competition: Schizophrenia classification challenge

2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), 2014

For the 24th Machine Learning for Signal Processing competition, participants were asked to autom... more For the 24th Machine Learning for Signal Processing competition, participants were asked to automatically diagnose schizophrenia using multimodal features derived from MRI scans. The objective of the classification task was to achieve the best possible schizophrenia diagnosis prediction based only on the multimodal features derived from brain MRI scans. A total of 2087 entries from 291 participants with active Kaggle.com accounts were made. Each participant developed a classifier, with optional feature selection, that combined functional and structural magnetic resonance imaging features. Here we review details about the competition setup, the winning strategies, and provide basic analyses of the submitted entries. We conclude with a discussion of the advances made to the neuroimaging and machine learning fields.

Download

Multidataset independent subspace analysis extends independent vector analysis

by Rogers F Silva and Sergey Plis

2014 IEEE International Conference on Image Processing (ICIP), 2014

ABSTRACT

Identification of Brain Image Biomarkers by Optimized Selection of Multimodal Independent Components

2008 IEEE Southwest Symposium on Image Analysis and Interpretation, 2008

The acquisition of multiple imaging modalities on the same individual in brain imaging studies ha... more The acquisition of multiple imaging modalities on the same individual in brain imaging studies has become a common practice. In functional studies, often several different tasks are performed on the same person. It is likely and intuitive that such datasets contain some shared and some unique information across the different acquired modalities. This motivates an analysis method that directly looks for the joint (shared) information lying within multimodal datasets. However, the cross-correlational standard-of-practice for examination of different regions does not easily allow for the study of the 'cross'-information of all brain voxels. A data fusion approach on the other hand has been shown to be beneficial in providing a unified framework for heterogenic datasets (Savopol et. al.). Hence the present work builds on the data fusion framework introduced by Calhoun et. al. called joint independent component analysis (jICA). As a 'second level' analysis this method utilizes summary maps of the patterns of brain activation generated for each individual in the dataset. Henceforth we refer to these summary maps as features derived from 'first-level' analysis of the original 'raw', 4D data using classic methods like the general linear model (GLM). Thus, with features generated for each modality and each individual of the dataset a jICA data fusion scheme as shown in is applied to yield joint (multimodal), maximally independent components (ICs) which capture the joint information from multiple modalities and enable identification of brain imaging biomarkers.

Download

Correction: Independent Component Analysis for Brain fMRI Does Indeed Select for Maximal Independence

by Rogers F Silva and Sergey Plis

PLoS ONE, 2013

A statistically motivated framework for simulation of stochastic data fusion models applied to multimodal neuroimaging

by Rogers F Silva and Sergey Plis

NeuroImage, 2014

Multimodal fusion is becoming more common as it proves to be a powerful approach to identify comp... more Multimodal fusion is becoming more common as it proves to be a powerful approach to identify complementary information from multimodal datasets. However, simulation of joint information is not straightforward. Published approaches mostly employ limited, provisional designs that often break the link between the model assumptions and the data for the sake of demonstrating properties of fusion techniques. This work introduces a new approach to synthetic data generation which allows full-compliance between data and model while still representing realistic spatiotemporal features in accordance with the current neuroimaging literature. The focus is on the simulation of joint information for the verification of stochastic linear models, particularly those used in multimodal data fusion of brain imaging data. Our first goal is to obtain a benchmark ground-truth in which estimation errors due to model mismatch are minimal or none. Then we move on to assess how estimation is affected by gradually increasing model discrepancies toward a more realistic dataset. The key aspect of our approach is that it permits complete control over the type and level of model mismatch, allowing for more educated inferences about the limitations and caveats of select stochastic linear models. As a result, impartial comparison of models is possible based on their performance in multiple different scenarios. Our proposed method uses the commonly overlooked theory of copulas to enable full control of the type and level of dependence/association between modalities, with no occurrence of spurious multimodal associations. Moreover, our approach allows for arbitrary single-modality marginal distributions for any fixed choice of dependence/association between multimodal features. Using our simulation framework, we can rigorously challenge the assumptions of several existing multimodal fusion approaches. Our study brings a new perspective to the problem of simulating multimodal data that can be used for ground-truth verification of various stochastic multimodal models available in the literature, and reveals some important aspects that are not captured or are overlooked by ad hoc simulations that lack a firm statistical motivation.

Divergence Measurements for the Optimal Identification of Multimodal Biomarkers

Download

Identification of multimodal MRI and EEG biomarkers using joint-ICA and divergence criteria

Motivated by Chernoff's bound on asymptotic probability of error we propose the alpha-divergence ... more Motivated by Chernoff's bound on asymptotic probability of error we propose the alpha-divergence measure and a surrogate, the alpha-Jensen difference, for indexing and retrieval in image and other databases. The alpha-divergence, also known as Renyi divergence, is a generalization of the Kullback-Liebler divergence and the Hellinger/Battacharya distance between the probability density characterizing image features of the query and the density characterizing features of candidates in the database. As in any divergence-based classification problem, the alpha-divergence must be estimated from the query or reference object and the objects in the database. The surrogate for the alpha-divergence, called the alpha-Jensen difference, can be simply estimated using non-parametric estimation of the joint alpha-entropy of the merged pairs of feature vectors. Two methods of alpha-entropy estimation are investigated: (1) indirect methods based on parametric or non-parametric density estimation over feature space; and (2) direct methods based on combinatorial optimization of minimal spanning trees or other continuous quasi-additive graphs over feature space. We analyze convergence rates and establish that the bias convergence rates of the MST entropy estimator can be better than that of an indirect estimator implemented with minimax adaptive kernel density estimation. We illustrate the MST estimator for geo-registration of images.

Download