A Bayesian approach for spatially adaptive regularisation in non-rigid registration
This paper introduces a novel method for inferring spatially varying regularisation in non-rigid registration. This is achieved through full Bayesian inference on a probabilistic registration model, where the prior on transformations is parametrised as a weighted mixture of spatially localised compo...
| Principais autores: | , , , , , , |
|---|---|
| Formato: | Journal article |
| Idioma: | English |
| Publicado em: |
2013
|
| _version_ | 1826298012617408512 |
|---|---|
| author | Simpson, I Woolrich, M Cardoso, M Cash, D Modat, M Schnabel, J Ourselin, S |
| author_facet | Simpson, I Woolrich, M Cardoso, M Cash, D Modat, M Schnabel, J Ourselin, S |
| author_sort | Simpson, I |
| collection | OXFORD |
| description | This paper introduces a novel method for inferring spatially varying regularisation in non-rigid registration. This is achieved through full Bayesian inference on a probabilistic registration model, where the prior on transformations is parametrised as a weighted mixture of spatially localised components. Such an approach has the advantage of allowing the registration to be more flexibly driven by the data than a more traditional global regularisation scheme, such as bending energy. The proposed method adaptively determines the influence of the prior in a local region. The importance of the prior may be reduced in areas where the data better supports deformations, or can enforce a stronger constraint in less informative areas. Consequently, the use of such a spatially adaptive prior may reduce the unwanted impact of regularisation on the inferred deformation field. This is especially important for applications such as tensor based morphometry, where the features of interest are directly derived from the deformation field. The proposed approach is demonstrated with application to tensor based morphometry analysis of subjects with Alzheimer's disease and healthy controls. The results show that using the proposed spatially adaptive prior leads to deformation fields that have a substantially lower average complexity, but which also provide more accurate localisation of statistical group differences. © 2013 Springer-Verlag. |
| first_indexed | 2024-03-07T04:40:20Z |
| format | Journal article |
| id | oxford-uuid:d16759cb-3641-4ca9-8923-009b7190b590 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T04:40:20Z |
| publishDate | 2013 |
| record_format | dspace |
| spelling | oxford-uuid:d16759cb-3641-4ca9-8923-009b7190b5902022-03-27T07:56:47ZA Bayesian approach for spatially adaptive regularisation in non-rigid registrationJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:d16759cb-3641-4ca9-8923-009b7190b590EnglishSymplectic Elements at Oxford2013Simpson, IWoolrich, MCardoso, MCash, DModat, MSchnabel, JOurselin, SThis paper introduces a novel method for inferring spatially varying regularisation in non-rigid registration. This is achieved through full Bayesian inference on a probabilistic registration model, where the prior on transformations is parametrised as a weighted mixture of spatially localised components. Such an approach has the advantage of allowing the registration to be more flexibly driven by the data than a more traditional global regularisation scheme, such as bending energy. The proposed method adaptively determines the influence of the prior in a local region. The importance of the prior may be reduced in areas where the data better supports deformations, or can enforce a stronger constraint in less informative areas. Consequently, the use of such a spatially adaptive prior may reduce the unwanted impact of regularisation on the inferred deformation field. This is especially important for applications such as tensor based morphometry, where the features of interest are directly derived from the deformation field. The proposed approach is demonstrated with application to tensor based morphometry analysis of subjects with Alzheimer's disease and healthy controls. The results show that using the proposed spatially adaptive prior leads to deformation fields that have a substantially lower average complexity, but which also provide more accurate localisation of statistical group differences. © 2013 Springer-Verlag. |
| spellingShingle | Simpson, I Woolrich, M Cardoso, M Cash, D Modat, M Schnabel, J Ourselin, S A Bayesian approach for spatially adaptive regularisation in non-rigid registration |
| title | A Bayesian approach for spatially adaptive regularisation in non-rigid registration |
| title_full | A Bayesian approach for spatially adaptive regularisation in non-rigid registration |
| title_fullStr | A Bayesian approach for spatially adaptive regularisation in non-rigid registration |
| title_full_unstemmed | A Bayesian approach for spatially adaptive regularisation in non-rigid registration |
| title_short | A Bayesian approach for spatially adaptive regularisation in non-rigid registration |
| title_sort | bayesian approach for spatially adaptive regularisation in non rigid registration |
| work_keys_str_mv | AT simpsoni abayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT woolrichm abayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT cardosom abayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT cashd abayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT modatm abayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT schnabelj abayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT ourselins abayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT simpsoni bayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT woolrichm bayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT cardosom bayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT cashd bayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT modatm bayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT schnabelj bayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration AT ourselins bayesianapproachforspatiallyadaptiveregularisationinnonrigidregistration |