Sergey Plis

This paper explores privacy-preserving visualization techniques for brain functional connectivity using differential privacy to protect participant data while maintaining visual utility.

This paper introduces a method to estimate directed causal brain networks from fMRI data, showing it detects more group-differentiating edges in schizophrenia compared to existing methods.

This paper introduces st-DenseViT, a weakly supervised spatiotemporal vision transformer that generates 4D dynamic brain networks from fMRI data, showing significant differences between schizophrenia patients and healthy controls.

This study uses an ODE model to separate rest and task signals in fMRI data, showing task-specific dynamics are subsets of rest dynamics and improving reaction time predictions.

The paper proposes a decentralized sparse federated learning approach for neuroimaging data to reduce communication overhead while maintaining data privacy.

This paper predicts mental health quality using deep learning on resting-state fMRI data, achieving higher accuracy than alternative models.

This paper uses Neural ODE to estimate initial conditions for brain network models to synchronize simulations with fMRI data, showing agreement between short-term and long-term parameterizations.

This paper introduces COINSTAC Vaults to enhance collaborative neuroimaging research through federated analysis and reproducibility by hosting standardized datasets.

This study uses federated analysis in COINSTAC to examine functional network connectivity in adolescent brains linked to smoking and alcohol consumption, finding hypoconnectivity and reduced synchronization.

This work proposes a decentralized method for brain age estimation using MRI data and shows it achieves similar performance to centralized models.

The paper introduces decentralized quality control tools for neuroimaging data in federated environments, showing promising results in visualization and outlier detection while preserving privacy.

This paper uses COINSTAC to perform decentralized VBM analysis on multi-site neuroimaging data to identify age, BMI, and smoking-related structural brain changes.

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.

This paper proposes a classification-based method to determine the optimal number of states in dynamic functional connectivity analysis, showing improved classification accuracy compared to traditional methods.

This paper introduces a statelet-based approach to analyze dynamic functional network connectivity, finding reduced modularity and lower connection consistency in schizophrenia patients compared to healthy controls.

The paper detects and corrects label noise in mental health diagnoses using MRI data from mood and psychosis disorders through a multi-label data cleansing approach.

The paper introduces a decentralized classification method that reduces network load while maintaining accuracy, applied to brain imaging data.

This paper applies deep residual networks to predict Alzheimer’s disease progression from mild cognitive impairment using neuroimaging data, achieving better performance than existing methods.