FMRI Classification

fMRI classification aims to decode brain activity patterns from fMRI data to identify cognitive states or diagnose neurological disorders. Current research heavily utilizes deep learning, particularly 3D convolutional neural networks (CNNs) and transformer architectures, often incorporating techniques like multi-pooling, time-distributed layers, and self-supervised pretraining to improve classification accuracy. These advancements are improving the ability to analyze fMRI data for applications such as identifying brain states associated with visual processing or diagnosing conditions like ADHD, offering potential for more accurate and objective clinical assessments. Furthermore, graph-based methods are emerging to leverage functional connectivity information for improved population-level analysis.

Papers

May 15, 2023

Self-Supervised Pretraining on Paired Sequences of fMRI Data for Transfer Learning to Brain Decoding Tasks
Sean Paulsen, Michael Casey
Transfer Learning Functional Magnetic Resonance Imaging Sequence of Sequence Self Supervised Pretraining Brain Decoding Auditory Cortex FMRI Classification

March 25, 2023

Multi-pooling 3D Convolutional Neural Network for fMRI Classification of Visual Brain States
Zhen Zhang, Masaki Takeda, Makoto Iwata
Convolutional Neural Network 3D Convolutional 3D Convolution Neural Decoding FMRI Classification

May 24, 2022

Highly Accurate FMRI ADHD Classification using time distributed multi modal 3D CNNs
Christopher Sims
Time Matter Functional Magnetic Resonance Imaging fMRI Analysis 3D Generative Adversarial Network FMRI Classification

March 26, 2022

Contrastive Graph Learning for Population-based fMRI Classification
Xuesong Wang, Lina Yao, Islem Rekik, Yu Zhang
Graph Contrastive Learning Contrastive Representation Learning Functional Connectivity FMRI Classification

FMRI Classification

Papers

Self-Supervised Pretraining on Paired Sequences of fMRI Data for Transfer Learning to Brain Decoding Tasks

Multi-pooling 3D Convolutional Neural Network for fMRI Classification of Visual Brain States

Highly Accurate FMRI ADHD Classification using time distributed multi modal 3D CNNs

Contrastive Graph Learning for Population-based fMRI Classification