Adaptive Smoothing in fMRI Data Processing Neural Networks
This work addresses the problem of optimizing fMRI data processing for neuroscientists and clinicians by enabling adaptive smoothing, though it is incremental as it builds on existing end-to-end learning paradigms.
The paper tackled the suboptimal spatial smoothing step in fMRI data processing pipelines by introducing an adaptive smoothing layer within end-to-end learning networks, achieving improved brain activity detection in finger tapping tasks with real fMRI data.
Functional Magnetic Resonance Imaging (fMRI) relies on multi-step data processing pipelines to accurately determine brain activity; among them, the crucial step of spatial smoothing. These pipelines are commonly suboptimal, given the local optimisation strategy they use, treating each step in isolation. With the advent of new tools for deep learning, recent work has proposed to turn these pipelines into end-to-end learning networks. This change of paradigm offers new avenues to improvement as it allows for a global optimisation. The current work aims at benefitting from this paradigm shift by defining a smoothing step as a layer in these networks able to adaptively modulate the degree of smoothing required by each brain volume to better accomplish a given data analysis task. The viability is evaluated on real fMRI data where subjects did alternate between left and right finger tapping tasks.