Limits...
Evaluation of Cross-Protocol Stability of a Fully Automated Brain Multi-Atlas Parcellation Tool.

Liang Z, He X, Ceritoglu C, Tang X, Li Y, Kutten KS, Oishi K, Miller MI, Mori S, Faria AV - PLoS ONE (2015)

Bottom Line: The entire brain was parceled into five different levels of granularity.Our results indicated that, with proper pre-processing steps, the impact of different protocols is minor compared to biological effects, such as age and pathology.A precise knowledge of the sources of data variation enables sufficient statistical power and ensures the reliability of an anatomical analysis when using this automated brain parcellation tool on datasets from various imaging protocols, such as clinical databases.

View Article: PubMed Central - PubMed

Affiliation: College of Electronics and Information Engineering, Sichuan University, Chengdu, China; The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, Maryland, United States of America.

ABSTRACT
Brain parcellation tools based on multiple-atlas algorithms have recently emerged as a promising method with which to accurately define brain structures. When dealing with data from various sources, it is crucial that these tools are robust for many different imaging protocols. In this study, we tested the robustness of a multiple-atlas, likelihood fusion algorithm using Alzheimer's Disease Neuroimaging Initiative (ADNI) data with six different protocols, comprising three manufacturers and two magnetic field strengths. The entire brain was parceled into five different levels of granularity. In each level, which defines a set of brain structures, ranging from eight to 286 regions, we evaluated the variability of brain volumes related to the protocol, age, and diagnosis (healthy or Alzheimer's disease). Our results indicated that, with proper pre-processing steps, the impact of different protocols is minor compared to biological effects, such as age and pathology. A precise knowledge of the sources of data variation enables sufficient statistical power and ensures the reliability of an anatomical analysis when using this automated brain parcellation tool on datasets from various imaging protocols, such as clinical databases.

No MeSH data available.


Related in: MedlinePlus

Brain parcellation scheme of the JHU multiple atlases.Multiple granularity levels (L1 to L5) are shown. Level 5 (L5) has the highest granularity and defines 286 regions. An anatomy-based hierarchical relationship was established to generate super-structures and lower-granularity parcellation, as shown in L1–L4.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4514626&req=5

pone.0133533.g001: Brain parcellation scheme of the JHU multiple atlases.Multiple granularity levels (L1 to L5) are shown. Level 5 (L5) has the highest granularity and defines 286 regions. An anatomy-based hierarchical relationship was established to generate super-structures and lower-granularity parcellation, as shown in L1–L4.

Mentions: Three-dimension (3D), magnetization-prepared, rapid gradient-echo (MPRAGE) sagittal images were used to evaluate brain morphology. The images were parcellated by an automated pipeline deployed at https://www.mricloud.org/, a platform that is open to the public after registration. The pipeline uses large diffeomorphic deformation metric mapping (LDDMM) and multi-atlas likelihood fusion (MALF) algorithms [24,26,27]. Twenty-three atlases (JHU adult atlas Version 5J) were used, in which the 286 structures were defined with a five-level ontological hierarchical relationship, as described by [25]. For the multiple-granularity analysis, the 286 defined structures were combined hierarchically to generate five different levels of anatomical representations. The numbers of the structures defined at each level were 8, 19, 53, 125, and 286 for levels 1 to 5, respectively (Fig 1). This pipeline accepts raw images without pre-processing, such as skull-stripping, matrix size, and orientation adjustment, intensity matching, or homogeneity correction. After the parcellation, the volumes of the defined structures were quantified.


Evaluation of Cross-Protocol Stability of a Fully Automated Brain Multi-Atlas Parcellation Tool.

Liang Z, He X, Ceritoglu C, Tang X, Li Y, Kutten KS, Oishi K, Miller MI, Mori S, Faria AV - PLoS ONE (2015)

Brain parcellation scheme of the JHU multiple atlases.Multiple granularity levels (L1 to L5) are shown. Level 5 (L5) has the highest granularity and defines 286 regions. An anatomy-based hierarchical relationship was established to generate super-structures and lower-granularity parcellation, as shown in L1–L4.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4514626&req=5

pone.0133533.g001: Brain parcellation scheme of the JHU multiple atlases.Multiple granularity levels (L1 to L5) are shown. Level 5 (L5) has the highest granularity and defines 286 regions. An anatomy-based hierarchical relationship was established to generate super-structures and lower-granularity parcellation, as shown in L1–L4.
Mentions: Three-dimension (3D), magnetization-prepared, rapid gradient-echo (MPRAGE) sagittal images were used to evaluate brain morphology. The images were parcellated by an automated pipeline deployed at https://www.mricloud.org/, a platform that is open to the public after registration. The pipeline uses large diffeomorphic deformation metric mapping (LDDMM) and multi-atlas likelihood fusion (MALF) algorithms [24,26,27]. Twenty-three atlases (JHU adult atlas Version 5J) were used, in which the 286 structures were defined with a five-level ontological hierarchical relationship, as described by [25]. For the multiple-granularity analysis, the 286 defined structures were combined hierarchically to generate five different levels of anatomical representations. The numbers of the structures defined at each level were 8, 19, 53, 125, and 286 for levels 1 to 5, respectively (Fig 1). This pipeline accepts raw images without pre-processing, such as skull-stripping, matrix size, and orientation adjustment, intensity matching, or homogeneity correction. After the parcellation, the volumes of the defined structures were quantified.

Bottom Line: The entire brain was parceled into five different levels of granularity.Our results indicated that, with proper pre-processing steps, the impact of different protocols is minor compared to biological effects, such as age and pathology.A precise knowledge of the sources of data variation enables sufficient statistical power and ensures the reliability of an anatomical analysis when using this automated brain parcellation tool on datasets from various imaging protocols, such as clinical databases.

View Article: PubMed Central - PubMed

Affiliation: College of Electronics and Information Engineering, Sichuan University, Chengdu, China; The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, Maryland, United States of America.

ABSTRACT
Brain parcellation tools based on multiple-atlas algorithms have recently emerged as a promising method with which to accurately define brain structures. When dealing with data from various sources, it is crucial that these tools are robust for many different imaging protocols. In this study, we tested the robustness of a multiple-atlas, likelihood fusion algorithm using Alzheimer's Disease Neuroimaging Initiative (ADNI) data with six different protocols, comprising three manufacturers and two magnetic field strengths. The entire brain was parceled into five different levels of granularity. In each level, which defines a set of brain structures, ranging from eight to 286 regions, we evaluated the variability of brain volumes related to the protocol, age, and diagnosis (healthy or Alzheimer's disease). Our results indicated that, with proper pre-processing steps, the impact of different protocols is minor compared to biological effects, such as age and pathology. A precise knowledge of the sources of data variation enables sufficient statistical power and ensures the reliability of an anatomical analysis when using this automated brain parcellation tool on datasets from various imaging protocols, such as clinical databases.

No MeSH data available.


Related in: MedlinePlus