arXiv subjects

Roberto C. Sotero

Publications and source records attributed to Roberto C. Sotero.

4 recordsLinked to original sources

arXiv2024

A Machine Learning Approach for Identifying Anatomical Biomarkers of Early Mild Cognitive Impairment

Alzheimer Disease poses a significant challenge, necessitating early detection for effective intervention. MRI is a key neuroimaging tool due to its ease of use and cost effectiveness. This study analyzes machine learning methods for MRI based biomarker selection and classification to distinguish between healthy controls and those who develop mild cognitive impairment within five years. Using 3 Tesla MRI data from ADNI and OASIS 3, we applied various machine learning techniques, including MATLAB Classification Learner app, nested cross validation, and Bayesian optimization. Data harmonization with polynomial regression improved performance. Consistent features identified were the entorhinal, hippocampus, lateral ventricle, and lateral orbitofrontal regions. For balanced ADNI data, Naive Bayes with z score harmonization performed best. For balanced OASIS 3, SVM with z score correction excelled. In imbalanced data, RUSBoost showed strong performance on ADNI and OASIS 3. Z score harmonization highlighted the potential of a semi automatic pipeline for early AD detection using MRI.

q-bio.NC↗

arXiv2024

Towards improving Alzheimer's intervention: a machine learning approach for biomarker detection through combining MEG and MRI pipelines

MEG are non invasive neuroimaging techniques with excellent temporal and spatial resolution, crucial for studying brain function in dementia and Alzheimer Disease. They identify changes in brain activity at various Alzheimer stages, including preclinical and prodromal phases. MEG may detect pathological changes before clinical symptoms, offering potential biomarkers for intervention. This study evaluates classification techniques using MEG features to distinguish between healthy controls and mild cognitive impairment participants from the BioFIND study. We compare MEG based biomarkers with MRI based anatomical features, both independently and combined. We used 3 Tesla MRI and MEG data from 324 BioFIND participants;158 MCI and 166 HC. Analyses were performed using MATLAB with SPM12 and OSL toolboxes. Machine learning analyses, including 100 Monte Carlo replications of 10 fold cross validation, were conducted on sensor and source spaces. Combining MRI with MEG features achieved the best performance; 0.76 accuracy and AUC of 0.82 for GLMNET using LCMV source based MEG. MEG only analyses using LCMV and eLORETA also performed well, suggesting that combining uncorrected MEG with z-score-corrected MRI features is optimal.

cs.LG↗

arXiv2023

Solving large-scale MEG/EEG source localization and functional connectivity problems simultaneously using state-space models

State-space models are widely employed across various research disciplines to study unobserved dynamics. Conventional estimation techniques, such as Kalman filtering and expectation maximisation, offer valuable insights but incur high computational costs in large-scale analyses. Sparse inverse covariance estimators can mitigate these costs, but at the expense of a trade-off between enforced sparsity and increased estimation bias, necessitating careful assessment in low signal-to-noise ratio (SNR) situations. To address these challenges, we propose a three-fold solution: 1) Introducing multiple penalised state-space (MPSS) models that leverage data-driven regularisation; 2) Developing novel algorithms derived from backpropagation, gradient descent, and alternating least squares to solve MPSS models; 3) Presenting a K-fold cross-validation extension for evaluating regularisation parameters. We validate this MPSS regularisation framework through lower and more complex simulations under varying SNR conditions, including a large-scale synthetic magneto- and electro-encephalography (MEG/EEG) data analysis. In addition, we apply MPSS models to concurrently solve brain source localisation and functional connectivity problems for real event-related MEG/EEG data, encompassing thousands of sources on the cortical surface. The proposed methodology overcomes the limitations of existing approaches, such as constraints to small-scale and region-of-interest analyses. Thus, it may enable a more accurate and detailed exploration of cognitive brain functions.

stat.CO↗

arXiv2019

Cross-frequency interactions during diffusion on complex brain networks are facilitated by scale-free properties

We studied the interactions between different temporal scales of diffusion processes on complex networks and found them to be stronger in scale-free (SF) than in Erdos-Renyi (ER) networks, especially for the case of phase-amplitude coupling (PAC)-the phenomenon where the phase of an oscillatory mode modulates the amplitude of another oscillation. We found that SF networks facilitate PAC between slow and fast frequency components of the diffusion process, whereas ER networks enable PAC between slow-frequency components. Nodes contributing the most to the generation of PAC in SF networks were non-hubs that connected with high probability to hubs. Additionally, brain networks from healthy controls (HC) and Alzheimer's disease (AD) patients presented a weaker PAC between slow and fast frequencies than SF, but higher than ER. We found that PAC decreased in AD compared to HC and was more strongly correlated to the scores of two different cognitive tests than what the strength of functional connectivity was, suggesting a link between cognitive impairment and multi-scale information flow in the brain.

q-bio.NC↗