The Automatic Neuroscientist: A framework for optimizing experimental design with closed-loop real-time fMRI

Romy Lorenz; Ricardo Pio Monti; Inês R. Violante; Christoforos Anagnostopoulos; Aldo A. Faisal; Giovanni Montana; Robert Leech

doi:10.1016/j.neuroimage.2016.01.032

The Automatic Neuroscientist: A framework for optimizing experimental design with closed-loop real-time fMRI

Romy Lorenz, Ricardo Pio Monti, Inês R. Violante, Christoforos Anagnostopoulos, Aldo A. Faisal, Giovanni Montana, Robert Leech^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

58 Citations (Scopus)

Abstract

Functional neuroimaging typically explores how a particular task activates a set of brain regions. Importantly though, the same neural system can be activated by inherently different tasks. To date, there is no approach available that systematically explores whether and how distinct tasks probe the same neural system. Here, we propose and validate an alternative framework, the Automatic Neuroscientist, which turns the standard fMRI approach on its head. We use real-time fMRI in combination with modern machine-learning techniques to automatically design the optimal experiment to evoke a desired target brain state. In this work, we present two proof-of-principle studies involving perceptual stimuli. In both studies optimization algorithms of varying complexity were employed; the first involved a stochastic approximation method while the second incorporated a more sophisticated Bayesian optimization technique. In the first study, we achieved convergence for the hypothesized optimum in 11 out of 14 runs in less than 10 min. Results of the second study showed how our closed-loop framework accurately and with high efficiency estimated the underlying relationship between stimuli and neural responses for each subject in one to two runs: with each run lasting 6.3 min. Moreover, we demonstrate that using only the first run produced a reliable solution at a group-level. Supporting simulation analyses provided evidence on the robustness of the Bayesian optimization approach for scenarios with low contrast-to-noise ratio. This framework is generalizable to numerous applications, ranging from optimizing stimuli in neuroimaging pilot studies to tailoring clinical rehabilitation therapy to patients and can be used with multiple imaging modalities in humans and animals.

Original language	English
Pages (from-to)	320-334
Number of pages	15
Journal	NeuroImage
Volume	129
DOIs	https://doi.org/10.1016/j.neuroimage.2016.01.032
Publication status	Published - 1 Apr 2016

Keywords

Bayesian optimization
Brain-computer interface
Closed-loop
Experimental design
Machine learning
Real-time fMRI

Access to Document

10.1016/j.neuroimage.2016.01.032

Cite this

@article{626333b25ba54d479918b661568c3d7d,

title = "The Automatic Neuroscientist: A framework for optimizing experimental design with closed-loop real-time fMRI",

abstract = "Functional neuroimaging typically explores how a particular task activates a set of brain regions. Importantly though, the same neural system can be activated by inherently different tasks. To date, there is no approach available that systematically explores whether and how distinct tasks probe the same neural system. Here, we propose and validate an alternative framework, the Automatic Neuroscientist, which turns the standard fMRI approach on its head. We use real-time fMRI in combination with modern machine-learning techniques to automatically design the optimal experiment to evoke a desired target brain state. In this work, we present two proof-of-principle studies involving perceptual stimuli. In both studies optimization algorithms of varying complexity were employed; the first involved a stochastic approximation method while the second incorporated a more sophisticated Bayesian optimization technique. In the first study, we achieved convergence for the hypothesized optimum in 11 out of 14 runs in less than 10 min. Results of the second study showed how our closed-loop framework accurately and with high efficiency estimated the underlying relationship between stimuli and neural responses for each subject in one to two runs: with each run lasting 6.3 min. Moreover, we demonstrate that using only the first run produced a reliable solution at a group-level. Supporting simulation analyses provided evidence on the robustness of the Bayesian optimization approach for scenarios with low contrast-to-noise ratio. This framework is generalizable to numerous applications, ranging from optimizing stimuli in neuroimaging pilot studies to tailoring clinical rehabilitation therapy to patients and can be used with multiple imaging modalities in humans and animals.",

keywords = "Bayesian optimization, Brain-computer interface, Closed-loop, Experimental design, Machine learning, Real-time fMRI",

author = "Romy Lorenz and Monti, {Ricardo Pio} and Violante, {In{\^e}s R.} and Christoforos Anagnostopoulos and Faisal, {Aldo A.} and Giovanni Montana and Robert Leech",

year = "2016",

month = apr,

day = "1",

doi = "10.1016/j.neuroimage.2016.01.032",

language = "English",

volume = "129",

pages = "320--334",

journal = "NeuroImage",

issn = "1053-8119",

publisher = "ACADEMIC PRESS INC ELSEVIER SCIENCE",

}

TY - JOUR

T1 - The Automatic Neuroscientist

T2 - A framework for optimizing experimental design with closed-loop real-time fMRI

AU - Lorenz, Romy

AU - Monti, Ricardo Pio

AU - Violante, Inês R.

AU - Anagnostopoulos, Christoforos

AU - Faisal, Aldo A.

AU - Montana, Giovanni

AU - Leech, Robert

PY - 2016/4/1

Y1 - 2016/4/1

N2 - Functional neuroimaging typically explores how a particular task activates a set of brain regions. Importantly though, the same neural system can be activated by inherently different tasks. To date, there is no approach available that systematically explores whether and how distinct tasks probe the same neural system. Here, we propose and validate an alternative framework, the Automatic Neuroscientist, which turns the standard fMRI approach on its head. We use real-time fMRI in combination with modern machine-learning techniques to automatically design the optimal experiment to evoke a desired target brain state. In this work, we present two proof-of-principle studies involving perceptual stimuli. In both studies optimization algorithms of varying complexity were employed; the first involved a stochastic approximation method while the second incorporated a more sophisticated Bayesian optimization technique. In the first study, we achieved convergence for the hypothesized optimum in 11 out of 14 runs in less than 10 min. Results of the second study showed how our closed-loop framework accurately and with high efficiency estimated the underlying relationship between stimuli and neural responses for each subject in one to two runs: with each run lasting 6.3 min. Moreover, we demonstrate that using only the first run produced a reliable solution at a group-level. Supporting simulation analyses provided evidence on the robustness of the Bayesian optimization approach for scenarios with low contrast-to-noise ratio. This framework is generalizable to numerous applications, ranging from optimizing stimuli in neuroimaging pilot studies to tailoring clinical rehabilitation therapy to patients and can be used with multiple imaging modalities in humans and animals.

AB - Functional neuroimaging typically explores how a particular task activates a set of brain regions. Importantly though, the same neural system can be activated by inherently different tasks. To date, there is no approach available that systematically explores whether and how distinct tasks probe the same neural system. Here, we propose and validate an alternative framework, the Automatic Neuroscientist, which turns the standard fMRI approach on its head. We use real-time fMRI in combination with modern machine-learning techniques to automatically design the optimal experiment to evoke a desired target brain state. In this work, we present two proof-of-principle studies involving perceptual stimuli. In both studies optimization algorithms of varying complexity were employed; the first involved a stochastic approximation method while the second incorporated a more sophisticated Bayesian optimization technique. In the first study, we achieved convergence for the hypothesized optimum in 11 out of 14 runs in less than 10 min. Results of the second study showed how our closed-loop framework accurately and with high efficiency estimated the underlying relationship between stimuli and neural responses for each subject in one to two runs: with each run lasting 6.3 min. Moreover, we demonstrate that using only the first run produced a reliable solution at a group-level. Supporting simulation analyses provided evidence on the robustness of the Bayesian optimization approach for scenarios with low contrast-to-noise ratio. This framework is generalizable to numerous applications, ranging from optimizing stimuli in neuroimaging pilot studies to tailoring clinical rehabilitation therapy to patients and can be used with multiple imaging modalities in humans and animals.

KW - Bayesian optimization

KW - Brain-computer interface

KW - Closed-loop

KW - Experimental design

KW - Machine learning

KW - Real-time fMRI

UR - http://www.scopus.com/inward/record.url?scp=84957078157&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2016.01.032

DO - 10.1016/j.neuroimage.2016.01.032

M3 - Article

C2 - 26804778

AN - SCOPUS:84957078157

SN - 1053-8119

VL - 129

SP - 320

EP - 334

JO - NeuroImage

JF - NeuroImage

ER -

The Automatic Neuroscientist: A framework for optimizing experimental design with closed-loop real-time fMRI

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this