fMRI Classification of Cognitive States Across Multiple Subjects
نویسنده
چکیده
With the evolvement of fMRI’s, a great amount of attention has been given to classifying cognitive states of human beings. Several machine learning approaches have been used to train single-subject classifiers to do so. We present a different method using a neural network and a RBF SVM to train one classifier across all subjects. For the single-subject classifier case, we experiment with PCA as a feature selection step, as well as train a neural network and a RBF SVM on the preprocessed data to compare its performance to previous results.
منابع مشابه
Training fMRI Classifiers to Detect Cognitive States across Multiple Human Subjects
We consider learning to classify cognitive states of human subjects, based on their brain activity observed via functional Magnetic Resonance Imaging (fMRI). This problem is important because such classifiers constitute “virtual sensors” of hidden cognitive states, which may be useful in cognitive science research and clinical applications. In recent work, Mitchell, et al. [6,7,9] have demonstr...
متن کاملTraining fMRI Classifiers to Discriminate Cognitive States across Multiple Subjects
We consider learning to classify cognitive states of human subjects, based on their brain activity observed via functional Magnetic Resonance Imaging (fMRI). This problem is important because such classifiers constitute “virtual sensors” of hidden cognitive states, which may be useful in cognitive science research and clinical applications. In recent work, Mitchell, et al. [6,7,9] have demonstr...
متن کاملPredictive fMRI Analysis for Multiple Subjects and Multiple Studies (Thesis)
In the context of predictive fMRI data analysis, the state of the art is to perform the analysis separately for each particular subject in a specific study. Given the nature of the fMRI data where there are many more features than instances, this kind of analysis might produce suboptimal predictive models since the data might not be sufficient to obtain accurate models. Based on findings in the...
متن کاملDetecting Cognitive States Using Machine Learning
Very little is known about the relationship between the cognitive states and the fMRI data, and very little is known about the feasibility of training classifiers to decode cognitive states. Our efforts aimed to automatically discover which spatial-temporal patterns in the fMRI data indicate a subject is performing a specific cognitive task, such as watching a picture or sentence. We developed ...
متن کاملFeature Selection for fMRI Classification Across Multiple Human Subjects
This paper investigates the use of fMRI data to develop a classifier to identify a subject’s cognitive state during a particular time interval. In particular, data from a set of subjects is used to decode the cognitive state of a new subject not used in the training process. This is a difficult task because each subject may produce different activation for a particular task and each has a diffe...
متن کامل