Differentiable programming for functional connectomics

Mapping the functional connectome has the potential to uncover key insights into brain organisation. However, existing workflows for functional connectomics are limited in their adaptability to new data, and principled workflow design is a challenging combinatorial problem. We introduce a new analytic paradigm and software toolbox that implements common operations used in functional connectomics as fully differentiable processing blocks. Under this paradigm, workflow configurations exist as reparameterisations of a differentiable functional that interpolates them. The differentiable program that we envision occupies a niche midway between traditional pipelines and end-to-end neural networks, combining the glass-box tractability and domain knowledge of the former with the amenability to optimisation of the latter. In this preliminary work, we provide a proof of concept for differentiable connectomics, demonstrating the capacity of our processing blocks both to recapitulate canonical knowledge in neuroscience and to make new discoveries in an unsupervised setting. Our differentiable modules are competitive with state-of-the-art methods in problem domains including functional parcellation, denoising, and covariance modelling. Taken together, our results and software demonstrate the promise of differentiable programming for functional connectomics.

• Publication year

  2022

• Venue

  ML4H@NeurIPS

• Publication date

  2022-05-31

• Fields of study

  Biology, Computer Science

• Identifiers
  DOI 10.48550/arXiv.2206.00649arXiv 2206.00649
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Dissociable multi-scale patterns of development in personalized brain networks

  2021cited by this paper
• A Comprehensive Survey on Community Detection With Deep Learning

  2021cited by this paper
• Deep learning encodes robust discriminative neuroimaging representations to outperform standard machine learning

  2021cited by this paper
• Neuroimaging: into the Multiverse

  2020cited by this paper
• Different scaling of linear models and deep learning in UKBiobank brain images versus machine-learning datasets

  2020cited by this paper
• Robust dynamic community detection with applications to human brain functional networks

  2020cited by this paper
• Topographic organization of the human subcortex unveiled with functional connectivity gradients

  2020cited by this paper
• Classifying Autism Spectrum Disorder Using the Temporal Statistics of Resting-State Functional MRI Data With 3D Convolutional Neural Networks

  2020cited by this paper
• Fine-grain atlases of functional modes for fMRI analysis

  2020cited by this paper
• Deep neural networks and kernel regression achieve comparable accuracies for functional connectivity prediction of behavior and demographics

  2020cited by this paper
• Optimising network modelling methods for fMRI

  2019cited by this paper
• The identification of temporal communities through trajectory clustering correlates with single-trial behavioural fluctuations in neuroimaging data

  2019cited by this paper
• Global Signal Regression Strengthens Association between Resting-State Functional Connectivity and Behavior

  2019influential reference
• Benchmarking functional connectome‐based predictive models for resting‐state fMRI

  2019cited by this paper
• High Performance I/O For Large Scale Deep Learning

  2019cited by this paper
• Variability in the analysis of a single neuroimaging dataset by many teams

  2019cited by this paper
• PyTorch: An Imperative Style, High-Performance Deep Learning Library

  2019cited by this paper
• AutoML: A Survey of the State-of-the-Art

  2019cited by this paper
• Brain state expression and transitions are related to complex executive cognition in normative neurodevelopment

  2018cited by this paper
• Correction of respiratory artifacts in MRI head motion estimates

  2018cited by this paper
• Averaging Weights Leads to Wider Optima and Better Generalization

  2018cited by this paper
• Imaging-based parcellations of the human brain

  2018cited by this paper
• Local-Global Parcellation of the Human Cerebral Cortex From Intrinsic Functional Connectivity MRI

  2017influential reference
• Sources and implications of whole‐brain fMRI signals in humans

  2017cited by this paper
• Optimizing fMRI preprocessing pipelines for block‐design tasks as a function of age

  2017cited by this paper
• An evaluation of the efficacy, reliability, and sensitivity of motion correction strategies for resting-state functional MRI

  2017influential reference
• Insight and inference for DVARS

  2017cited by this paper
• Benchmarking of participant‐level confound regression strategies for the control of motion artifact in studies of functional connectivity

  2017influential reference
• Precision Functional Mapping of Individual Human Brains.

  2017influential reference
• Towards end-to-end optimisation of functional image analysis pipelines

  2016cited by this paper
• A multi-modal parcellation of human cerebral cortex

  2016influential reference
• Generation and Evaluation of a Cortical Area Parcellation from Resting-State Correlations.

  2016cited by this paper
• Dynamic fluctuations coincide with periods of high and low modularity in resting-state functional brain networks

  2015cited by this paper
• Functional System and Areal Organization of a Highly Sampled Individual Human Brain.

  2015cited by this paper
• Evaluation of ICA-AROMA and alternative strategies for motion artifact removal in resting state fMRI

  2015cited by this paper
• Transport on Riemannian Manifold for Functional Connectivity-Based Classification

  2014cited by this paper
• MSM: A new flexible framework for Multimodal Surface Matching

  2014cited by this paper
• Adam: A Method for Stochastic Optimization

  2014cited by this paper
• Methods to detect, characterize, and remove motion artifact in resting state fMRI

  2014influential reference
• An improved framework for confound regression and filtering for control of motion artifact in the preprocessing of resting-state functional connectivity data

  2013influential reference
• The nuisance of nuisance regression: Spectral misspecification in a common approach to resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity

  2013influential reference
• Functional connectomics from resting-state fMRI.

  2013cited by this paper
• The minimal preprocessing pipelines for the Human Connectome Project

  2013influential reference
• Supplementary Material to: Tracking Whole-Brain Connectivity Dynamics in the Resting State.

  2013cited by this paper
• The WU-Minn Human Connectome Project: An overview

  2013cited by this paper
• On the Plurality of (Methodological) Worlds: Estimating the Analytic Flexibility of fMRI Experiments

  2012cited by this paper
• Impact of in-scanner head motion on multiple measures of functional connectivity: Relevance for studies of neurodevelopment in youth

  2012cited by this paper
• The Human Connectome Project: A data acquisition perspective

  2012influential reference
• Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion

  2012influential reference
• The organization of the human cerebral cortex estimated by intrinsic functional connectivity

  2011cited by this paper
• The impact of global signal regression on resting state correlations: Are anti-correlated networks introduced?

  2009cited by this paper
• Group comparison of resting-state FMRI data using multi-subject ICA and dual regression

  2009cited by this paper
• Fast unfolding of communities in large networks

  2008cited by this paper
• A component based noise correction method (CompCor) for BOLD and perfusion based fMRI

  2007cited by this paper
• Finding and evaluating community structure in networks.

  2003influential reference
• Neurophysiological investigation of the basis of the fMRI signal

  2001cited by this paper
• Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system.

  1999cited by this paper
• Least-squares frequency analysis of unequally spaced data

  1976cited by this paper

• No citing papers are available for this paper.