Over the past decade functional Magnetic Resonance Imaging (fMRI) has emerged as a powerful technique to locate activity of human brain while engaged in a particular task or cognitive state. We consider the inverse problem of detecting the cognitive state of a human subject based on the fMRI data. We have explored classification techniques such as Gaussian Naive Bayes, k-Nearest Neighbour and Support Vector Machines. In order to reduce the very high dimensional fMRI data, we have used three feature selection strategies. Discriminating features and activity based features were used to select features for the problem of identifying the instantaneous cognitive state given a single fMRI scan and correlation based features were used when fMRI data from a single time interval was given. A case study of visuo-motor sequence learning is presented. The set of cognitive states we are interested in detecting are whether the subject has learnt a sequence, and if the subject is paying attention only towards the position or towards both the color and position of the visual stimuli. We have successfully used correlation based features to detect position-color related cognitive states with 80 % accuracy and the cognitive states related to learning with 62.5 % accuracy. 1

We present a continuous and unobtrusive approach to analyze and reason about users ’ personal experiences of interacting with virtual and game environments. Focusing on an immersive educational game environment that we are developing, this is achieved through the capture and storage of user’s movements and events that occur as a result of interactions with and within immersive environments. Termed immersidata, we then query and analyze immersidata to make sense of user behavior. Two example approaches are described. The first describes an application ISIS (Immersidata analySIS) that provides a tool for analysis of user behavior/experience through the indexing of immersidata with video clips of students’ gaming sessions. This approach is described by way of an example to identify the causes of interruptions or breaks in interactions/focus of attention to facilitate the identification of problematic design. In our second example we describe our work towards classifying students ’ performance through immersidata. To this aim, we describe one example of transforming immersidata into multivariate time series and then by applying feature subset selection techniques we identify the features that differentiate students. We describe the application of this approach to identify novice and expert players with 90 % accuracy. One proposal is to use this to customize the game environment appropriate to the students’ ability. Finally, we present future directions for the continuation of the work presented herein and also, the application of the immersidata system to capture, store and analyze personal behavior/experiences and provide appropriate feedback in our work and home environments.

We present an approach for learning models that obtain accurate classification of data objects, collected in large scale spatiotemporal domains. The model generation is structured in three phases: spatial dimension reduction, spatiotemporal features extraction, and feature selection. Novel techniques for the first two phases are presented, with two alternatives for the middle phase. We explore model generation based on the combinations of techniques from each phase. We apply the introduced methodology to datasets from the Voltage-Sensitive Dye Imaging (VSDI) domain, where the resulting classification models successfully decode neuronal population responses in the visual cortex of behaving animals. VSDI is currently the best technique enabling simultaneous high spatial (10, 000 points) and temporal (10 ms or less) resolution imaging from neuronal population in the cortex. We demonstrate that not only our approach is scalable enough to handle computationally challenging data, but it also contributes to the neuroimaging field of study with its decoding abilities. The effectiveness of our methodology is further explored on a dataset from the hurricanes domain, and a promising direction, based on the preliminary results of hurricane severity classification, is revealed. Acknowledgments

Abstract—We present an approach for learning models that obtain accurate classification of large scale data objects, collected in spatiotemporal domains. The model generation is structured in three phases: pixel selection (spatial dimension reduction), spatiotemporal features extraction and feature selection. Novel techniques for the first two phases are presented, with two alternatives for the middle phase. Model generation based on the combinations of techniques from each phase is explored. The introduced methodology is applied on datasets from the Voltage-Sensitive Dye Imaging (VSDI) domain, where the generated classification models successfully decode neuronal population responses in the visual cortex of behaving animals. VSDI currently is the best technique enabling simultaneous high spatial (10,000 points) and temporal (10 ms or less) resolution imaging from neuronal population in the cortex. We demonstrate that not only our approach is scalable enough to handle computationally challenging data, but it also contributes to the neuroimaging field of study with its decoding abilities. Index Terms—classification; spatiotemporal; application; brain imaging; neural decoding; visual cortex; I.

Abstract—We present an approach for learning models that obtain accurate classification of large scale data objects, collected in spatiotemporal domains. The model generation is structured in three phases: pixel selection (spatial dimension reduction), spatiotemporal features extraction and feature selection. Novel techniques for the first two phases are presented, with two alternatives for the middle phase. Model generation based on the combinations of techniques from each phase is explored. The introduced methodology is applied on datasets from the Voltage-Sensitive Dye Imaging (VSDI) domain, where the generated classification models successfully decode neuronal population responses in the visual cortex of behaving animals. VSDI currently is the best technique enabling simultaneous high spatial (10,000 points) and temporal (10 ms or less) resolution imaging from neuronal population in the cortex. We demonstrate that not only our approach is scalable enough to handle computationally challenging data, but it also contributes to the neuroimaging field of study with its decoding abilities. Index Terms—classification; spatiotemporal; application; brain imaging; neural decoding; visual cortex; I.