• Corpus ID: 245650468

Learning shared neural manifolds from multi-subject FMRI data

@article{Huang2022LearningSN,
  title={Learning shared neural manifolds from multi-subject FMRI data},
  author={Je-chun Huang and Erica L. Busch and Tom Wallenstein and Michal Gerasimiuk and Andrew Benz and Guillaume Lajoie and Guy Wolf and Nicholas B. Turk-Browne and Smita Krishnaswamy},
  journal={ArXiv},
  year={2022},
  volume={abs/2201.00622}
}
Functional magnetic resonance imaging (fMRI) is a notoriously noisy measurement of brain activity because of the large variations between individuals, signals marred by environmental differences during collection, and spatiotemporal averaging required by the measurement resolution. In addition, the data is extremely high dimensional, with the space of the activity typically having much lower intrinsic dimension. In order to understand the connection between stimuli of interest and brain… 

Figures and Tables from this paper

Multi-view manifold learning of human brain state trajectories

1 Brain activity as measured with functional magnetic resonance imaging (fMRI) gives 2 the illusion of intractably high dimensionality, rife with collection and biological noise. 3 Nonlinear

Temporal PHATE: A multi-view manifold learning method for brain state trajectories

TLDR
T-PHATE demonstrates impressive improvements over previous cutting-edge approaches to understanding the nature of cognition from fMRI and bodes potential applications broadly for high-dimensional datasets of temporally-diffuse processes.

"Task-relevant autoencoding"enhances machine learning for human neuroscience

TLDR
A Task-Relevant Autoencoder via Classifier Enhancement (TRACE) was developed and tested its ability to extract behaviorally-relevant, separable representations compared to a standard autoencoding for two severely truncated machine learning datasets, showing up to 30% increased classification accuracy and up to threefold improvement in discovering “cleaner”, task-relevant representations.

References

SHOWING 1-10 OF 33 REFERENCES

Non-linear manifold learning in fMRI uncovers a low-dimensional space of brain dynamics

TLDR
It is established that a shared, robust, and interpretable low-dimensional space of brain dynamics can be recovered from a rich repertoire of task based fMRI data, and that resting-state data embeds fully onto the same task embedding, indicating similar brain states are present in both task and resting- state data.

Uncovering the Topology of Time-Varying fMRI Data using Cubical Persistence

TLDR
A novel topological approach is presented that encodes each time point in an fMRI data set as a persistence diagram of topological features, i.e. high-dimensional voids present in the data, which can be clustered to find meaningful groupings between participants and useful in studying within-subject brain state trajectories of subjects performing a particular task.

A high-resolution 7-Tesla fMRI dataset from complex natural stimulation with an audio movie

Here we present a high-resolution functional magnetic resonance (fMRI) dataset – 20 participants recorded at high field strength (7 Tesla) during prolonged stimulation with an auditory feature film

Group-PCA for very large fMRI datasets

A Reduced-Dimension fMRI Shared Response Model

TLDR
A shared response model for aggregating multi-subject fMRI data that accounts for different functional topographies among anatomically aligned datasets is developed and demonstrates improved sensitivity in identifying a shared response for a variety of datasets and anatomical brain regions of interest.

Extendable and invertible manifold learning with geometry regularized autoencoders

TLDR
This work presents a new method for integrating both approaches to manifold learning and autoencoders by incorporating a geometric regularization term in the bottleneck of the autoencoder, based on the diffusion potential distances from the recently-proposed PHATE visualization method.

Physiological Noise in Brainstem fMRI

TLDR
This Methods Article will provide a practical introduction to the techniques used to correct for the presence of physiological noise in time series fMRI data, and advice on modeling noise sources is given.