Limits...
A feature-selective independent component analysis method for functional MRI.

Li YO, Adali T, Calhoun VD - Int J Biomed Imaging (2007)

Bottom Line: In this work, we propose a simple and effective scheme to incorporate prior knowledge about the sources of interest (SOIs) in independent component analysis (ICA) and apply the method to estimate brain activations from functional magnetic resonance imaging (fMRI) data.The feature-selective scheme is achieved through a filtering operation in the source sample space followed by a projection onto the demixing vector space by a least squares projection in an iterative ICA process.We also compare the task-related sources estimated from true fMRI data by a feature-selective ICA algorithm versus an ICA algorithm and show evidence that the feature-selective scheme helps improve the estimation of the sources in both spatial activation patterns and the time courses.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science and Electrical Engineering, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250, USA.

ABSTRACT
In this work, we propose a simple and effective scheme to incorporate prior knowledge about the sources of interest (SOIs) in independent component analysis (ICA) and apply the method to estimate brain activations from functional magnetic resonance imaging (fMRI) data. We name the proposed method as feature-selective ICA since it incorporates the features in the sample space of the independent components during ICA estimation. The feature-selective scheme is achieved through a filtering operation in the source sample space followed by a projection onto the demixing vector space by a least squares projection in an iterative ICA process. We perform ICA estimation of artificial activations superimposed into a resting state fMRI dataset to show that the feature-selective scheme improves the detection of injected activation from the independent component estimated by ICA. We also compare the task-related sources estimated from true fMRI data by a feature-selective ICA algorithm versus an ICA algorithm and show evidence that the feature-selective scheme helps improve the estimation of the sources in both spatial activation patterns and the time courses.

No MeSH data available.


Related in: MedlinePlus

ROC plots for detection of the superimposed activation inthe hybrid fMRI data with FS-Infomax incorporating a priori activation templates overlapping withthe true activation region to different degrees (–), and ROC plot for detection of thesuperimposed activation with Infomax estimation.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2233814&req=5

fig2: ROC plots for detection of the superimposed activation inthe hybrid fMRI data with FS-Infomax incorporating a priori activation templates overlapping withthe true activation region to different degrees (–), and ROC plot for detection of thesuperimposed activation with Infomax estimation.

Mentions: Figure 2 shows the ROC curves for detection of the imposedactivation using FS-Infomax with apriori activation templates overlapping with the true activation to differentdegrees. It is observed that the Infomax algorithm with feature-selective schemeimproves the estimation of the superimposed activation by incorporating a priori location information about theactivation. The degree of improvement decreases as the prior deviates from theground truth. When the a prioritemplate poorly matches the true activation, that is, overlaps at ,the detection performance is decreased to the same level as the Infomaxalgorithm. It is worth noting that for the range of , which is usually of more practicalinterest for detection performance, FS-Infomax with all a priori templates show significantimprovement on detection power compared to Infomax. The results indicate thatthe feature-selective scheme is effective to the detection of activation atrelatively low CNR conditions.


A feature-selective independent component analysis method for functional MRI.

Li YO, Adali T, Calhoun VD - Int J Biomed Imaging (2007)

ROC plots for detection of the superimposed activation inthe hybrid fMRI data with FS-Infomax incorporating a priori activation templates overlapping withthe true activation region to different degrees (–), and ROC plot for detection of thesuperimposed activation with Infomax estimation.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2233814&req=5

fig2: ROC plots for detection of the superimposed activation inthe hybrid fMRI data with FS-Infomax incorporating a priori activation templates overlapping withthe true activation region to different degrees (–), and ROC plot for detection of thesuperimposed activation with Infomax estimation.
Mentions: Figure 2 shows the ROC curves for detection of the imposedactivation using FS-Infomax with apriori activation templates overlapping with the true activation to differentdegrees. It is observed that the Infomax algorithm with feature-selective schemeimproves the estimation of the superimposed activation by incorporating a priori location information about theactivation. The degree of improvement decreases as the prior deviates from theground truth. When the a prioritemplate poorly matches the true activation, that is, overlaps at ,the detection performance is decreased to the same level as the Infomaxalgorithm. It is worth noting that for the range of , which is usually of more practicalinterest for detection performance, FS-Infomax with all a priori templates show significantimprovement on detection power compared to Infomax. The results indicate thatthe feature-selective scheme is effective to the detection of activation atrelatively low CNR conditions.

Bottom Line: In this work, we propose a simple and effective scheme to incorporate prior knowledge about the sources of interest (SOIs) in independent component analysis (ICA) and apply the method to estimate brain activations from functional magnetic resonance imaging (fMRI) data.The feature-selective scheme is achieved through a filtering operation in the source sample space followed by a projection onto the demixing vector space by a least squares projection in an iterative ICA process.We also compare the task-related sources estimated from true fMRI data by a feature-selective ICA algorithm versus an ICA algorithm and show evidence that the feature-selective scheme helps improve the estimation of the sources in both spatial activation patterns and the time courses.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science and Electrical Engineering, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250, USA.

ABSTRACT
In this work, we propose a simple and effective scheme to incorporate prior knowledge about the sources of interest (SOIs) in independent component analysis (ICA) and apply the method to estimate brain activations from functional magnetic resonance imaging (fMRI) data. We name the proposed method as feature-selective ICA since it incorporates the features in the sample space of the independent components during ICA estimation. The feature-selective scheme is achieved through a filtering operation in the source sample space followed by a projection onto the demixing vector space by a least squares projection in an iterative ICA process. We perform ICA estimation of artificial activations superimposed into a resting state fMRI dataset to show that the feature-selective scheme improves the detection of injected activation from the independent component estimated by ICA. We also compare the task-related sources estimated from true fMRI data by a feature-selective ICA algorithm versus an ICA algorithm and show evidence that the feature-selective scheme helps improve the estimation of the sources in both spatial activation patterns and the time courses.

No MeSH data available.


Related in: MedlinePlus