Decoding Cognitive States Using the Bag of Words Model on fMRI Time Series

Date

2016-05-19

Author

Sucu, Güneş
Akbaş, Emre
Yarman Vural, Fatoş Tunay

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

57
views

0
downloads

Bag-of-words (BoW) modeling has yielded successful results in document and image classification tasks. In this paper, we explore the use of BoW for cognitive state classification. We estimate a set of common patterns embedded in the fMRI time series recorded in three dimensional voxel coordinates by clustering the BOLD responses. We use these common patterns, called the code-words, to encode activities of both individual voxels and group of voxels, and obtain a BoW representation on which we train linear classifiers. Our experimental results show that the BoW encoding, when applied to individual voxels, significantly improves the classification accuracy (an average 7.2% increase over 13 different datasets) compared to a classical multi voxel pattern analysis method. This preliminary result gives us a clue to generate a code-book for fMRI data which may be used to represent a variety of cognitive states to study the human brain.

Subject Keywords

Time series analysis, Histograms, Brain modeling , Encoding, Training, Feature extraction, Computational modeling

URI

https://hdl.handle.net/11511/37381

DOI

https://doi.org/10.1109/siu.2016.7496222

Conference Name

24th Signal Processing and Communication Application Conference (SIU)

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Decoding cognitive states using the bag of words model on fMRI time series Sucu, Güneş; Akbaş, Emre; Department of Computer Engineering (2017) Bag-of-words (BoW) modeling has yielded successful results in document and image classification tasks. In this study, we explore the use of BoW for cognitive state classification. We estimate a set of common patterns embedded in the Functional Magnetic Resonance Imaging (fMRI) time series recorded in three dimensional voxel coordinates by clustering the Blood Oxygen Level Dependent (BOLD) responses. We use these common patterns, called the code-words, to encode activities of both individual voxels and group...
Representation Learning for Contextual Object and Region Detection in Remote Sensing Firat, Orhan; Can, Gulcan; Yarman Vural, Fatoş Tunay (2014-08-28) The performance of object recognition and classification on remote sensing imagery is highly dependent on the quality of extracted features, amount of labelled data and the priors defined for contextual models. In this study, we examine the representation learning opportunities for remote sensing. First we attacked localization of contextual cues for complex object detection using disentangling factors learnt from a small amount of labelled data. The complex object, which consists of several sub-parts is fu...
Utilization of feature modeling in axiomatic design Üçtepe, Orhan; Doğru, Ali Hikmet; Department of Computer Engineering (2008) This thesis provides an approach to use feature modeling with a set of guidelines for requirements definition and decomposition activities of the axiomatic design methodology. A tool that supports the development of feature models and modeling of the Axiomatic Design activities is implemented to be utilized for guiding the designer. Axiomatic Design suggested four domains of information in the transformation of the problem definition to the solution, and provided mechanisms for supporting the mapping among ...
An Improved BOW Approach Using Fuzzy Feature Encoding and Visual-word Weighting Altintakan, Umit L.; Yazıcı, Adnan (2015-08-05) The bag-of-words (BOW) has become a popular image representation model with successful implementations in visual analysis. Although the original model has been improved in several ways, the utilization of the Fuzzy Set Theory in BOW has not been investigated thoroughly. This paper presents a fuzzy feature encoding approach to address the problems associated with the hard and soft assignments of image features to the visualwords. Our encoding method assigns each image feature to only the first and second clo...
Component extraction analysis of multivariate time series Akman, I; DeGooijer, JG (1996-05-01) A method for modelling several observed parallel time series is proposed. The method involves seeking possible common underlying pure AR and MA components in the series. The common components are forced to be mutually uncorrelated so that univariate time series modelling and forecasting techniques can be applied. The proposed method is shown to be a useful addition to the time series analyst's toolkit, if common sources of variation in multivariate data need to be quickly identified.

Citation Formats

G. Sucu, E. Akbaş, and F. T. Yarman Vural, “Decoding Cognitive States Using the Bag of Words Model on fMRI Time Series,” presented at the 24th Signal Processing and Communication Application Conference (SIU), 2016, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/37381.