Search Results for author: Sergey Plis

Found 25 papers, 5 papers with code

DynaLay: An Introspective Approach to Dynamic Layer Selection for Deep Networks

no code implementations20 Dec 2023 Mrinal Mathur, Sergey Plis

Deep learning models have become increasingly computationally intensive, requiring extensive computational resources and time for both training and inference.

Decision Making

Brainchop: Next Generation Web-Based Neuroimaging Application

1 code implementation24 Oct 2023 Mohamed Masoud, Pratyush Reddy, Farfalla Hu, Sergey Plis

Performing volumetric image processing directly within the browser, particularly with medical data, presents unprecedented challenges compared to conventional backend tools.

Learning low-dimensional dynamics from whole-brain data improves task capture

no code implementations18 May 2023 Eloy Geenjaar, Donghyun Kim, Riyasat Ohib, Marlena Duda, Amrit Kashyap, Sergey Plis, Vince Calhoun

We evaluate our approach on various task-fMRI datasets, including motor, working memory, and relational processing tasks, and demonstrate that it outperforms widely used dimensionality reduction techniques in how well the latent timeseries relates to behavioral sub-tasks, such as left-hand or right-hand tapping.

Dimensionality Reduction

SalientGrads: Sparse Models for Communication Efficient and Data Aware Distributed Federated Training

no code implementations15 Apr 2023 Riyasat Ohib, Bishal Thapaliya, Pratyush Gaggenapalli, Jingyu Liu, Vince Calhoun, Sergey Plis

Federated learning (FL) enables the training of a model leveraging decentralized data in client sites while preserving privacy by not collecting data.

Federated Learning

Self-Supervised Mental Disorder Classifiers via Time Reversal

no code implementations29 Nov 2022 Zafar Iqbal, Usman Mahmood, Zening Fu, Sergey Plis

In this paper, we demonstrate that a model trained on the time direction of functional neuro-imaging data could help in any downstream task, for example, classifying diseases from healthy controls in fMRI data.

Pipeline-Invariant Representation Learning for Neuroimaging

no code implementations27 Aug 2022 Xinhui Li, Alex Fedorov, Mrinal Mathur, Anees Abrol, Gregory Kiar, Sergey Plis, Vince Calhoun

We next propose two pipeline-invariant representation learning methodologies, MPSL and PXL, to improve robustness in classification performance and to capture similar neural network representations.

Representation Learning

Geometrically Guided Integrated Gradients

no code implementations13 Jun 2022 Md Mahfuzur Rahman, Noah Lewis, Sergey Plis

We demonstrate through extensive experiments that the proposed approach outperforms vanilla and integrated gradients in subjective and quantitative assessment.

Constraint-Based Causal Structure Learning from Undersampled Graphs

no code implementations18 May 2022 Mohammadsajad Abavisani, David Danks, Vince Calhoun, Sergey Plis

Graphical structures estimated by causal learning algorithms from time series data can provide highly misleading causal information if the causal timescale of the generating process fails to match the measurement timescale of the data.

Informativeness Time Series +1

Deep Dynamic Effective Connectivity Estimation from Multivariate Time Series

no code implementations4 Feb 2022 Usman Mahmood, Zening Fu, Vince Calhoun, Sergey Plis

To bridge this gap, we developed dynamic effective connectivity estimation via neural network training (DECENNT), a novel model to learn an interpretable directed and dynamic graph induced by the downstream classification/prediction task.

Connectivity Estimation Link Prediction +2

A deep learning model for data-driven discovery of functional connectivity

1 code implementation7 Dec 2021 Usman Mahmood, Zening Fu, Vince Calhoun, Sergey Plis

Functional connectivity (FC) studies have demonstrated the overarching value of studying the brain and its disorders through the undirected weighted graph of fMRI correlation matrix.

Multi network InfoMax: A pre-training method involving graph convolutional networks

no code implementations1 Nov 2021 Usman Mahmood, Zening Fu, Vince Calhoun, Sergey Plis

Since almost every DL model is an ensemble of multiple networks, we take our high-level embeddings from two different networks of a model --a convolutional and a graph network--.

Graph Classification

Brain dynamics via Cumulative Auto-Regressive Self-Attention

no code implementations1 Nov 2021 Usman Mahmood, Zening Fu, Vince Calhoun, Sergey Plis

The supervised training of the model as a classifier between patients and controls results in a model that generates directed connectivity graphs and highlights the components of the time-series that are predictive for each subject.

Time Series Time Series Analysis

Attend to connect: end-to-end brain functional connectivity estimation

no code implementations ICLR Workshop GTRL 2021 Usman Mahmood, Zening Fu, Vince Calhoun, Sergey Plis

Functional connectivity (FC) studies have demonstrated the benefits of investigating the brain and its disorders through the undirected weighted graph of fMRI correlation matrix.

Connectivity Estimation

Explicit Group Sparse Projection with Applications to Deep Learning and NMF

no code implementations9 Dec 2019 Riyasat Ohib, Nicolas Gillis, Niccolò Dalmasso, Sameena Shah, Vamsi K. Potluru, Sergey Plis

Instead, in our approach we set the sparsity level for the whole set explicitly and simultaneously project a group of vectors with the sparsity level of each vector tuned automatically.

Network Pruning

Transfer Learning of fMRI Dynamics

no code implementations16 Nov 2019 Usman Mahmood, Md Mahfuzur Rahman, Alex Fedorov, Zening Fu, Sergey Plis

In this paper, we demonstrate a self-supervised pre-training method that enables us to pre-train directly on fMRI dynamics of healthy control subjects and transfer the learning to much smaller datasets of schizophrenia.

Small Data Image Classification Transfer Learning

Prediction of Progression to Alzheimer's disease with Deep InfoMax

no code implementations24 Apr 2019 Alex Fedorov, R. Devon Hjelm, Anees Abrol, Zening Fu, Yuhui Du, Sergey Plis, Vince D. Calhoun

Arguably, unsupervised learning plays a crucial role in the majority of algorithms for processing brain imaging.

General Classification

Improving Classification Rate of Schizophrenia Using a Multimodal Multi-Layer Perceptron Model with Structural and Functional MR

no code implementations4 Apr 2018 Alvaro Ulloa, Sergey Plis, Vince Calhoun

We propose the use of a multimodal multi-layer perceptron model to enhance the predictive power of structural and functional magnetic resonance imaging (sMRI and fMRI) combined.

General Classification

Almost instant brain atlas segmentation for large-scale studies

1 code implementation1 Nov 2017 Alex Fedorov, Eswar Damaraju, Vince Calhoun, Sergey Plis

Complexity of the task increases even further when segmenting structural MRI of the brain into an atlas with more than 50 regions.

End-to-end learning of brain tissue segmentation from imperfect labeling

2 code implementations3 Dec 2016 Alex Fedorov, Jeremy Johnson, Eswar Damaraju, Alexei Ozerin, Vince Calhoun, Sergey Plis

Segmenting a structural magnetic resonance imaging (MRI) scan is an important pre-processing step for analytic procedures and subsequent inferences about longitudinal tissue changes.

Causal Discovery from Subsampled Time Series Data by Constraint Optimization

no code implementations25 Feb 2016 Antti Hyttinen, Sergey Plis, Matti Järvisalo, Frederick Eberhardt, David Danks

This paper focuses on causal structure estimation from time series data in which measurements are obtained at a coarser timescale than the causal timescale of the underlying system.

Causal Discovery Time Series +1

Rate-Agnostic (Causal) Structure Learning

no code implementations NeurIPS 2015 Sergey Plis, David Danks, Cynthia Freeman, Vince Calhoun

That is, these algorithms all learn causal structure without assuming any particular relation between the measurement and system timescales; they are thus rate-agnostic.

Time Series Time Series Analysis

Cannot find the paper you are looking for? You can Submit a new open access paper.