Improving intersubject image registration using tissue-class information benefits robustness and accuracy of multi-atlas based anatomical segmentation

Heckemann, Rolf A.; Keihaninejad, Shiva; Aljabar, Paul; Rueckert, Daniel; Hajnal, Joseph V.; Hammers, Alexander

doi:10.1016/j.neuroimage.2010.01.072

Cited by 178 publications

(168 citation statements)

References 25 publications

Supporting

Mentioning

162

Contrasting

Unclassified

Order By: Relevance

“…Different approaches have been reported in literature to select the most appropriate atlas for segmentation, either using a representative average atlas (47) or multiple atlases 44 , 48 . In the future, we hope to develop a selection algorithm that would select only atlases that show good anatomical matching as judged through the metric in the initial rigid alignment.…”

Section: Discussionmentioning

confidence: 99%

Multiatlas segmentation of thoracic and abdominal anatomy with level set‐based local search

Schreibmann

Marcus

Fox

2014

J Applied Clin Med Phys

View full text Add to dashboard Cite

Segmentation of organs at risk (OARs) remains one of the most time‐consuming tasks in radiotherapy treatment planning. Atlas‐based segmentation methods using single templates have emerged as a practical approach to automate the process for brain or head and neck anatomy, but pose significant challenges in regions where large interpatient variations are present. We show that significant changes are needed to autosegment thoracic and abdominal datasets by combining multi‐atlas deformable registration with a level set‐based local search. Segmentation is hierarchical, with a first stage detecting bulk organ location, and a second step adapting the segmentation to fine details present in the patient scan. The first stage is based on warping multiple presegmented templates to the new patient anatomy using a multimodality deformable registration algorithm able to cope with changes in scanning conditions and artifacts. These segmentations are compacted in a probabilistic map of organ shape using the STAPLE algorithm. Final segmentation is obtained by adjusting the probability map for each organ type, using customized combinations of delineation filters exploiting prior knowledge of organ characteristics. Validation is performed by comparing automated and manual segmentation using the Dice coefficient, measured at an average of 0.971 for the aorta, 0.869 for the trachea, 0.958 for the lungs, 0.788 for the heart, 0.912 for the liver, 0.884 for the kidneys, 0.888 for the vertebrae, 0.863 for the spleen, and 0.740 for the spinal cord. Accurate atlas segmentation for abdominal and thoracic regions can be achieved with the usage of a multi‐atlas and perstructure refinement strategy. To improve clinical workflow and efficiency, the algorithm was embedded in a software service, applying the algorithm automatically on acquired scans without any user interaction.PACS numbers: 87.57.nm, 87.57.N‐, 87.57.Q‐

show abstract

Section: Discussionmentioning

confidence: 99%

Multiatlas segmentation of thoracic and abdominal anatomy with level set‐based local search

Schreibmann

Marcus

Fox

2014

J Applied Clin Med Phys

View full text Add to dashboard Cite

show abstract

“…Magnetic resonance (MR) images were segmented into anatomical regions using multi-atlas label propagation with expectation-maximization (MALPEM) [29]. Here, 30 atlases 3 segmented into 134 anatomical regions were transformed to an unsegmented image space using non-rigid registration [38,19]. Individual atlas label maps were then transformed to the unsegmented image space using the calculated transformations and a nearest neighbor interpolation scheme.…”

Section: Datamentioning

confidence: 99%

Instantiated mixed effects modeling of Alzheimer's disease markers

et al. 2016

View full text Add to dashboard Cite

The assessment and prediction of a subject's current and future risk of developing neurodegenerative diseases like Alzheimer's disease is of great interest in both the design of clinical trials as well as in clinical decision making. Exploring the longitudinal trajectory of markers related to neurodegeneration is an important task when selecting subjects for treatment in trials and the clinic, in the evaluation of early disease indicators and the monitoring of disease progression. Given that there is substantial intersubject variability, models that attempt to describe marker trajectories for a whole population will likely lack specificity for the representation of individual patients. Therefore, we argue here that individualized models provide a more accurate alternative that can be used for tasks such as population stratification and a subject-specific prognosis. In the work presented here, mixed effects modeling is used to derive a global and individual marker trajectories for a training population. Test subject (new patient) specific models are then instantiated using a stratified "marker signature" that defines a subpopulation of similar * Corresponding author: Ricardo Guerrero Email address: reg09@imperial.ac.uk (R. Guerrero) 1 This project was partially funded by the Innovate UK (formerly Technology Strategy Board -TSB).2 Data used in the preparation of this article was obtained from the ADNI database (http://adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how to apply/ ADNI Acknowledgement List.pdf Preprint submitted to NeuroImage June 27, 2016 cases within the training database. From this subpopulation, personalized models of the expected trajectory of several markers are subsequently estimated for unseen patients. These patient specific models of markers are shown to provide better predictions of time-to-conversion to Alzheimer's disease than population based models.

show abstract

“…The proposed validation will thus have two components. First, the overlap accuracy of multilabel information propagation will be estimated and compared to MAPER [4] using a leave one out approach on the 30 young controls. Then, the accuracy of information extrapolation accuracy will be characterised by propagating the brain segmentations from the elderly control group to the MCI and AD patients.…”

Section: Validationmentioning

confidence: 99%

“…The accuracy of propagating information through a geodesic path was compared to MAPER [4], a direct information fusion method based on majority voting. As the amount of parcelations available for validation is limited, a leave-one-out cross validation was performed only on the 30 young controls that have manual brain parcelations.…”

Section: Multi-label Propagation Accuracymentioning

confidence: 99%

“…However, in the case of limited and morphologically clustered source of information, like the 30 young control subjects with an associated parcelation of 83 key brain areas provided by Hammers et al [3], structural parcelation propagation can be problematic. As these parcelations are defined only on young controls with normal anatomy, it is non trivial to directly map this information to morphologically dissimilar and pathological subjects [4,5] without introducing large errors. More recently, Wolz et al [6] introduced the LEAP approach (learning embeddings for atlas propagation) for brain segmentation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Geodesic Information Flows

Cardoso

Wolz

Modat

et al. 2012

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2012

Self Cite

View full text Add to dashboard Cite

Abstract. Homogenising the availability of manually generated information in large databases has been a key challenge of medical imaging for many years. Due to the time consuming nature of manually segmenting, parcellating and localising landmarks in medical images, these sources of information tend to be scarce and limited to small, and sometimes morphologically similar, subsets of data. In this work we explore a new framework where these sources of information can be propagated to morphologically dissimilar images by diffusing and mapping the information through intermediate steps. The spatially variant data embedding uses the local morphology and intensity similarity between images to diffuse the information only between locally similar images. This framework can thus be used to propagate any information from any group of subject to every other subject in a database with great accuracy. Comparison to state-of-the-art propagation methods showed highly statistically significant (p < 10 −4 ) improvements in accuracy when propagating both structural parcelations and brain segmentations geodesically.

show abstract

Improving intersubject image registration using tissue-class information benefits robustness and accuracy of multi-atlas based anatomical segmentation

Cited by 178 publications

References 25 publications

Multiatlas segmentation of thoracic and abdominal anatomy with level set‐based local search

Multiatlas segmentation of thoracic and abdominal anatomy with level set‐based local search

Instantiated mixed effects modeling of Alzheimer's disease markers

Geodesic Information Flows

Contact Info

Product

Resources

About