J. Järvinen, Ronald Borra, J. Kulmala, H. Aronen, A. Korvenoja, E. Salli
{"title":"多主体fMRI数据的上下文分析方法","authors":"J. Järvinen, Ronald Borra, J. Kulmala, H. Aronen, A. Korvenoja, E. Salli","doi":"10.1145/2093698.2093720","DOIUrl":null,"url":null,"abstract":"In this article, we present a novel approach to create single subject fMRI (functional magnetic resonance imaging) activation maps by utilizing the group data of several subjects on the individual level. Classification of a voxel as either activated or non-activated is based on its z-value and classification in the neighborhood. We defined the neighborhood of a voxel to consist of corresponding and neighboring voxels in two subjects in addition to the neighbourhood voxels within the subject itself. Determination of the two neighboring subjects was based on kappa-statistics between single subject activation maps calculated without data from other subjects. This approach was taken using multi-subject contextual clustering. Both fully simulated and real subject null data with simulated activations were used. ROC (receiver operator characteristics) analysis showed increased classification accuracy of activated and non-activated voxels when using the described approach.","PeriodicalId":91990,"journal":{"name":"... International Symposium on Applied Sciences in Biomedical and Communication Technologies. International Symposium on Applied Sciences in Biomedical and Communication Technologies","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2011-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A contextual analysis method for multi-subject fMRI data\",\"authors\":\"J. Järvinen, Ronald Borra, J. Kulmala, H. Aronen, A. Korvenoja, E. Salli\",\"doi\":\"10.1145/2093698.2093720\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this article, we present a novel approach to create single subject fMRI (functional magnetic resonance imaging) activation maps by utilizing the group data of several subjects on the individual level. Classification of a voxel as either activated or non-activated is based on its z-value and classification in the neighborhood. We defined the neighborhood of a voxel to consist of corresponding and neighboring voxels in two subjects in addition to the neighbourhood voxels within the subject itself. Determination of the two neighboring subjects was based on kappa-statistics between single subject activation maps calculated without data from other subjects. This approach was taken using multi-subject contextual clustering. Both fully simulated and real subject null data with simulated activations were used. ROC (receiver operator characteristics) analysis showed increased classification accuracy of activated and non-activated voxels when using the described approach.\",\"PeriodicalId\":91990,\"journal\":{\"name\":\"... International Symposium on Applied Sciences in Biomedical and Communication Technologies. International Symposium on Applied Sciences in Biomedical and Communication Technologies\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2011-10-26\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"... International Symposium on Applied Sciences in Biomedical and Communication Technologies. International Symposium on Applied Sciences in Biomedical and Communication Technologies\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2093698.2093720\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"... International Symposium on Applied Sciences in Biomedical and Communication Technologies. International Symposium on Applied Sciences in Biomedical and Communication Technologies","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2093698.2093720","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A contextual analysis method for multi-subject fMRI data
In this article, we present a novel approach to create single subject fMRI (functional magnetic resonance imaging) activation maps by utilizing the group data of several subjects on the individual level. Classification of a voxel as either activated or non-activated is based on its z-value and classification in the neighborhood. We defined the neighborhood of a voxel to consist of corresponding and neighboring voxels in two subjects in addition to the neighbourhood voxels within the subject itself. Determination of the two neighboring subjects was based on kappa-statistics between single subject activation maps calculated without data from other subjects. This approach was taken using multi-subject contextual clustering. Both fully simulated and real subject null data with simulated activations were used. ROC (receiver operator characteristics) analysis showed increased classification accuracy of activated and non-activated voxels when using the described approach.