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

Distribution of Bonferroni-corrected p-values for protocol differences in 286 structures at Level 5.For better visualization, the p-values (P, y-axis) are presented as Log10(P/5). A p-value of 0.05 corresponds to -2 on the y-axis. At this threshold, two regions reached statistical significance.
© Copyright Policy
Related In: Results  -  Collection

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

pone.0133533.g002: Distribution of Bonferroni-corrected p-values for protocol differences in 286 structures at Level 5.For better visualization, the p-values (P, y-axis) are presented as Log10(P/5). A p-value of 0.05 corresponds to -2 on the y-axis. At this threshold, two regions reached statistical significance.

Mentions: We examined protocol-dependent bias in the volume measurements at the five different granularity levels. The volumes of the defined structures did not show significant differences in any of the six protocols at the levels of granularity 1 to 4. At these levels, the lowest p-values were 0.5184, 0.3869, 0.6264, and 0.1565, respectively, suggesting an insignificant influence of the protocols. Fig 2 shows the distribution of Bonferroni-corrected p-values of the 286 structures at level 5. At this highest granularity level, two regions reached statistical significance. The actual regions and inter-protocol differences are shown in Fig 3. These regions were the white matter of the left inferior temporal (ITWM_L, p-value = 0.04) and right rectus gyrus (RGWM_R, p-value = 0.008), in which GE scanners tend to have smaller values. These regions constitute approximately 0.35% and 0.15% of the total brain volume, respectively.


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)

Distribution of Bonferroni-corrected p-values for protocol differences in 286 structures at Level 5.For better visualization, the p-values (P, y-axis) are presented as Log10(P/5). A p-value of 0.05 corresponds to -2 on the y-axis. At this threshold, two regions reached statistical significance.
© Copyright Policy
Related In: Results  -  Collection

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

pone.0133533.g002: Distribution of Bonferroni-corrected p-values for protocol differences in 286 structures at Level 5.For better visualization, the p-values (P, y-axis) are presented as Log10(P/5). A p-value of 0.05 corresponds to -2 on the y-axis. At this threshold, two regions reached statistical significance.
Mentions: We examined protocol-dependent bias in the volume measurements at the five different granularity levels. The volumes of the defined structures did not show significant differences in any of the six protocols at the levels of granularity 1 to 4. At these levels, the lowest p-values were 0.5184, 0.3869, 0.6264, and 0.1565, respectively, suggesting an insignificant influence of the protocols. Fig 2 shows the distribution of Bonferroni-corrected p-values of the 286 structures at level 5. At this highest granularity level, two regions reached statistical significance. The actual regions and inter-protocol differences are shown in Fig 3. These regions were the white matter of the left inferior temporal (ITWM_L, p-value = 0.04) and right rectus gyrus (RGWM_R, p-value = 0.008), in which GE scanners tend to have smaller values. These regions constitute approximately 0.35% and 0.15% of the total brain volume, respectively.

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