Fast Shape-Based Nearest-Neighbor Search for Brain MRIs Using Hierarchical Feature Matching

Zhu, Peihong; Awate, Suyash P.; Gerber, Samuel; Whitaker, Ross T.

doi:10.1007/978-3-642-23629-7_59

Cited by 7 publications

(7 citation statements)

References 13 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Aljabar et al (2008) derived image similarities from anatomical segmentation overlaps. Zhu et al (2011) find nearest neighbors by combining edge extraction with spatial pyramid matching. BrainPrint relates to these approaches because it provides a measure of brain similarity.…”

Section: Introductionmentioning

confidence: 99%

BrainPrint: A discriminative characterization of brain morphology

Golland²,

et al. 2015

View full text Add to dashboard Cite

We introduce BrainPrint, a compact and discriminative representation of brain morphology. BrainPrint captures shape information of an ensemble of cortical and subcortical structures by solving the eigenvalue problem of the 2D and 3D Laplace-Beltrami operator on triangular (boundary) and tetrahedral (volumetric) meshes. This discriminative characterization enables new ways to study the similarity between brains; the focus can either be on a specific brain structure of interest or on the overall brain similarity. We highlight four applications for BrainPrint in this article: (i) subject identification, (ii) age and sex prediction, (iii) brain asymmetry analysis, and (iv) potential genetic influences on brain morphology. The properties of BrainPrint require the derivation of new algorithms to account for the heterogeneous mix of brain structures with varying discriminative power. We conduct experiments on three datasets, including over 3000 MRI scans from the ADNI database, 436 MRI scans from the OASIS dataset, and 236 MRI scans from the VETSA twin study. All processing steps for obtaining the compact representation are fully automated, making this processing framework particularly attractive for handling large datasets.

show abstract

Section: Introductionmentioning

confidence: 99%

BrainPrint: A discriminative characterization of brain morphology

Golland²,

et al. 2015

View full text Add to dashboard Cite

show abstract

“…

The database can be organized using a hierarchical groupwise nonlinear-diffeomorphic registration scheme [31], [32] that produces several optimal mean templates for multiple classes of images within the database. A target can be mapped first to all the mean templates to determine the most similar classes and then the target can be registered to templates only within the similar classes.

We can use fast approximate searches for similar templates relying on affine registration followed by spatial pyramid matching on coded geometry-capturing features, e.g., canny edges clustered and coded based on orientation and curvature [23]. The approximate search can be used to select a number of templates larger than the required k *( M ) (e.g., twice or thrice) k *( M ), after which the selected templates can be nonlinearly registered to the target.…”

Section: Resultsmentioning

confidence: 99%

“…The experiments in this paper use an isotropic patch of size 21 3 mm 3 . The proposed framework indeed allows the usage of more sophisticated Mercer kernels such as the pyramid match kernel [21] and the spatial pyramid kernel [22] that was used for multiatlas segmentation in [13], [23]. …”

Section: Methodsmentioning

confidence: 99%

Multiatlas Segmentation as Nonparametric Regression

Awate

Whitaker

2014

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

This paper proposes a novel theoretical framework to model and analyze the statistical characteristics of a wide range of segmentation methods that incorporate a database of label maps or atlases; such methods are termed as label fusion or multiatlas segmentation. We model these multiatlas segmentation problems as nonparametric regression problems in the high-dimensional space of image patches. We analyze the nonparametric estimator’s convergence behavior that characterizes expected segmentation error as a function of the size of the multiatlas database. We show that this error has an analytic form involving several parameters that are fundamental to the specific segmentation problem (determined by the chosen anatomical structure, imaging modality, registration algorithm, and label-fusion algorithm). We describe how to estimate these parameters and show that several human anatomical structures exhibit the trends modeled analytically. We use these parameter estimates to optimize the regression estimator. We show that the expected error for large database sizes is well predicted by models learned on small databases. Thus, a few expert segmentations can help predict the database sizes required to keep the expected error below a specified tolerance level. Such cost-benefit analysis is crucial for deploying clinical multiatlas segmentation systems.

show abstract

“…However, multiatlas approaches require only a few most-similar templates ( k in k NN) and registration between the target and thousands of templates in a large database can be very expensive. Thus, this paper uses an extremely-fast approximate search for similar templates relying on affine registration followed by spatial pyramid matching on coded geometry-capturing features (canny edges clustered and coded based on orientation and curvature) [18]. This implicitly induces a distance metric in the space of deformed images f , underlying k NN regression.…”

Section: Experiments and Results On A Clinical Databasementioning

confidence: 99%

“…This has lead to multiatlas , nonparametric atlas, or label-fusion approaches [1,2,5,10,11,13,15,16] to segmentation that leverage information in the entire database of atlases. Multiatlas approaches can exploit methods for fast selection [18] of a small subset of templates that are most similar to the target. They independently register the selected templates to the target and, then, deform database segmentations to the target space.…”

Section: Introductionmentioning

confidence: 99%

How Many Templates Does It Take for a Good Segmentation?: Error Analysis in Multiatlas Segmentation as a Function of Database Size

Awate

Zhu

Whitaker

2012

Multimodal Brain Image Analysis

Self Cite

View full text Add to dashboard Cite

This paper proposes a novel formulation to model and analyze the statistical characteristics of some types of segmentation problems that are based on combining label maps / templates / atlases. Such segmentation-by-example approaches are quite powerful on their own for several clinical applications and they provide prior information, through spatial context, when combined with intensity-based segmentation methods. The proposed formulation models a class of multiatlas segmentation problems as nonparametric regression problems in the high-dimensional space of images. The paper presents a systematic analysis of the nonparametric estimation's convergence behavior (i.e. characterizing segmentation error as a function of the size of the multiatlas database) and shows that it has a specific analytic form involving several parameters that are fundamental to the specific segmentation problem (i.e. chosen anatomical structure, imaging modality, registration method, label-fusion algorithm, etc.). We describe how to estimate these parameters and show that several brain anatomical structures exhibit the trends determined analytically. The proposed framework also provides per-voxel confidence measures for the segmentation. We show that the segmentation error for large database sizes can be predicted using small-sized databases. Thus, small databases can be exploited to predict the database sizes required ("how many templates") to achieve "good" segmentations having errors lower than a specified tolerance. Such cost-benefit analysis is crucial for designing and deploying multiatlas segmentation systems.

show abstract

Fast Shape-Based Nearest-Neighbor Search for Brain MRIs Using Hierarchical Feature Matching

Cited by 7 publications

References 13 publications

BrainPrint: A discriminative characterization of brain morphology

BrainPrint: A discriminative characterization of brain morphology

Multiatlas Segmentation as Nonparametric Regression

How Many Templates Does It Take for a Good Segmentation?: Error Analysis in Multiatlas Segmentation as a Function of Database Size

Contact Info

Product

Resources

About