Rogers F. Silva

Multimodal subspace independent vector analysis effectively captures the latent relationships between brain structure and function

This paper introduces MSIVA, a method to capture latent relationships between brain structure and function, showing it identifies phenotype-associated brain regions in neuroimaging data.

Analysis of Longitudinal Change Patterns in Developing Brain Using Functional and Structural Magnetic Resonance Imaging via Multimodal Fusion

The paper links functional change patterns to brain structure using multimodal fusion in adolescents, revealing sex-specific differences in neural dynamics.

aNy-way ICA and its application to estimate cortico-thalamo-cerebellar functional links in schizophrenia

This paper uses aNy-way ICA to identify a cortico-thalamo-cerebellar circuit in schizophrenia, showing altered connectivity and cognitive impairments across multiple datasets.

An information theory framework for capturing multi-connectivity via spatial network encoding reveals reduced population count (Hamming weight) localized to auditory, visual, and motor networks

The paper introduces a voxel-centric framework using information theory to analyze multi-way brain network interactions, revealing reduced Hamming weights in auditory, visual, and motor networks in schizophrenia.

Direct linkage detection with multimodal IVA fusion reveals markers of age, sex, cognition, and schizophrenia in large neuroimaging studies

This paper develops a method to detect linked features across neuroimaging modalities and identifies markers related to age, sex, cognition, and schizophrenia in large datasets.

dcSBM: A federated constrained source-based morphometry approach for multivariate brain structure mapping

The paper introduces dcSBM, a decentralized constrained source-based morphometry method for multivariate brain structure analysis, showing it reliably identifies group differences in brain morphology across sites.

Individualized Spatial Network Predictions Using Siamese Convolutional Neural Networks: A Resting-State fMRI Study of over 11,000 Unaffected Individuals

This study uses a Siamese CNN framework to predict individual identity from resting-state fMRI data, achieving 88% accuracy in subject discrimination using spatial network pairs.

Hype versus hope: Deep learning encodes more predictive and robust brain imaging representations than standard machine learning

This study compares deep learning and standard machine learning on brain imaging data, showing deep learning's superior predictive power and robustness in age and gender classification.

Modular and state-relevant connectivity in high-frequency resting-state BOLD fMRI data: An independent component analysis

The paper analyzes resting-state fMRI data across different frequency bins to assess functional network connectivity and finds distinct patterns at higher frequencies.

Sparse parallel independent component analysis and its application to identify linked genomic and gray matter alterations underlying working memory impairment in attention-deficit/hyperactivity disorder

This paper introduces spICA to link genomic and brain imaging data, identifying a SNP and GMV association in ADHD linked to working memory deficits.

Parallel Group ICA + ICA: Joint Estimation of Linked Functional Network Variability and Structural Covariation with Application to Schizophrenia

This paper introduces a method to fuse functional and structural MRI data by integrating temporal information, identifying linked variability and covariation in schizophrenia.