Cluster extent inference revisited: quantification and localization of brain activity
Cluster inference based on spatial extent thresholding is a popular analysis method multiple testing in spatial data, and is frequently used for finding activated brain areas in neuroimaging. However, the method has several well-known issues. While powerful for finding regions with some activation,...
| Main Authors: | , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Royal Statistical Society
2023
|
| _version_ | 1797112148712751104 |
|---|---|
| author | Goeman, JJ Gorecki, P Monajemi, R Nichols, TE Weeda, W |
| author_facet | Goeman, JJ Gorecki, P Monajemi, R Nichols, TE Weeda, W |
| author_sort | Goeman, JJ |
| collection | OXFORD |
| description | Cluster inference based on spatial extent thresholding is a popular analysis method multiple testing in spatial data, and is frequently used for finding activated brain areas in neuroimaging. However, the method has several well-known issues. While powerful for finding regions with some activation, the method as currently defined does not allow any further quantification or localisation of signal. In this paper, we repair this gap. We show that cluster-extent inference can be used (1) to infer the presence of signal in any region of interest and (2) to quantify the percentage of activation in such regions. These additional inferences come for free, i.e. they do not require any further adjustment of the alpha-level of tests, while retaining full family-wise error control. We achieve this extension of the possibilities of cluster inference by embedding the method into a closed testing procedure, and solving the graph-theoretic k-separator problem that results from this embedding. We demonstrate the usefulness of the improved method in a large-scale application to neuroimaging data from the Neurovault database. |
| first_indexed | 2024-03-07T08:20:07Z |
| format | Journal article |
| id | oxford-uuid:66114f25-1987-4fb7-b5e2-2682e00193eb |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T08:20:07Z |
| publishDate | 2023 |
| publisher | Royal Statistical Society |
| record_format | dspace |
| spelling | oxford-uuid:66114f25-1987-4fb7-b5e2-2682e00193eb2024-01-24T09:26:57ZCluster extent inference revisited: quantification and localization of brain activityJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:66114f25-1987-4fb7-b5e2-2682e00193ebEnglishSymplectic ElementsRoyal Statistical Society2023Goeman, JJGorecki, PMonajemi, RNichols, TEWeeda, WCluster inference based on spatial extent thresholding is a popular analysis method multiple testing in spatial data, and is frequently used for finding activated brain areas in neuroimaging. However, the method has several well-known issues. While powerful for finding regions with some activation, the method as currently defined does not allow any further quantification or localisation of signal. In this paper, we repair this gap. We show that cluster-extent inference can be used (1) to infer the presence of signal in any region of interest and (2) to quantify the percentage of activation in such regions. These additional inferences come for free, i.e. they do not require any further adjustment of the alpha-level of tests, while retaining full family-wise error control. We achieve this extension of the possibilities of cluster inference by embedding the method into a closed testing procedure, and solving the graph-theoretic k-separator problem that results from this embedding. We demonstrate the usefulness of the improved method in a large-scale application to neuroimaging data from the Neurovault database. |
| spellingShingle | Goeman, JJ Gorecki, P Monajemi, R Nichols, TE Weeda, W Cluster extent inference revisited: quantification and localization of brain activity |
| title | Cluster extent inference revisited: quantification and localization of brain activity |
| title_full | Cluster extent inference revisited: quantification and localization of brain activity |
| title_fullStr | Cluster extent inference revisited: quantification and localization of brain activity |
| title_full_unstemmed | Cluster extent inference revisited: quantification and localization of brain activity |
| title_short | Cluster extent inference revisited: quantification and localization of brain activity |
| title_sort | cluster extent inference revisited quantification and localization of brain activity |
| work_keys_str_mv | AT goemanjj clusterextentinferencerevisitedquantificationandlocalizationofbrainactivity AT goreckip clusterextentinferencerevisitedquantificationandlocalizationofbrainactivity AT monajemir clusterextentinferencerevisitedquantificationandlocalizationofbrainactivity AT nicholste clusterextentinferencerevisitedquantificationandlocalizationofbrainactivity AT weedaw clusterextentinferencerevisitedquantificationandlocalizationofbrainactivity |