Limits...
Nonlocal intracranial cavity extraction.

Manjón JV, Eskildsen SF, Coupé P, Romero JE, Collins DL, Robles M - Int J Biomed Imaging (2014)

Bottom Line: However, the high variability of brain shape and size due to normal intersubject variability, normal changes occurring over the lifespan, and abnormal changes due to disease makes the ICV estimation problem challenging.To this end, an improved nonlocal label fusion scheme based on BEaST technique is proposed to increase the accuracy of the ICV estimation.The proposed method is compared with recent state-of-the-art methods and the results demonstrate an improved performance both in terms of accuracy and reproducibility while maintaining a reduced computational burden.

View Article: PubMed Central - PubMed

Affiliation: Instituto de Aplicaciones de las Tecnologías de la Información y de las Comunicaciones Avanzadas (ITACA), Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, Spain.

ABSTRACT
Automatic and accurate methods to estimate normalized regional brain volumes from MRI data are valuable tools which may help to obtain an objective diagnosis and followup of many neurological diseases. To estimate such regional brain volumes, the intracranial cavity volume (ICV) is often used for normalization. However, the high variability of brain shape and size due to normal intersubject variability, normal changes occurring over the lifespan, and abnormal changes due to disease makes the ICV estimation problem challenging. In this paper, we present a new approach to perform ICV extraction based on the use of a library of prelabeled brain images to capture the large variability of brain shapes. To this end, an improved nonlocal label fusion scheme based on BEaST technique is proposed to increase the accuracy of the ICV estimation. The proposed method is compared with recent state-of-the-art methods and the results demonstrate an improved performance both in terms of accuracy and reproducibility while maintaining a reduced computational burden.

No MeSH data available.


Related in: MedlinePlus

Evolution of segmentation accuracy in function of the number of training subject templates used in the segmentation process.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4195262&req=5

fig4: Evolution of segmentation accuracy in function of the number of training subject templates used in the segmentation process.

Mentions: To study the impact of the method parameters, an exhaustive search of the optimum values was performed using the LOO dataset using the library segmentations as gold standard references. Each one of the 49 subjects in the library was processed using the remaining cases of the library as priors and the resulting segmentation was compared to the manual labeling. To measure segmentation accuracy, the Dice coefficient [38] was used. Method parameters such as patch size and search area were set as in BEaST method while an exhaustive search for the optimal number of templates N used for the segmentation process was carried out (see Figure 4). This search demonstrated that the segmentation accuracy stabilizes around N = [20–30] range which is in good agreement with previous results from BEaST. We decided to use N = 30 as default value given the reduced computational cost of the proposed method.


Nonlocal intracranial cavity extraction.

Manjón JV, Eskildsen SF, Coupé P, Romero JE, Collins DL, Robles M - Int J Biomed Imaging (2014)

Evolution of segmentation accuracy in function of the number of training subject templates used in the segmentation process.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4195262&req=5

fig4: Evolution of segmentation accuracy in function of the number of training subject templates used in the segmentation process.
Mentions: To study the impact of the method parameters, an exhaustive search of the optimum values was performed using the LOO dataset using the library segmentations as gold standard references. Each one of the 49 subjects in the library was processed using the remaining cases of the library as priors and the resulting segmentation was compared to the manual labeling. To measure segmentation accuracy, the Dice coefficient [38] was used. Method parameters such as patch size and search area were set as in BEaST method while an exhaustive search for the optimal number of templates N used for the segmentation process was carried out (see Figure 4). This search demonstrated that the segmentation accuracy stabilizes around N = [20–30] range which is in good agreement with previous results from BEaST. We decided to use N = 30 as default value given the reduced computational cost of the proposed method.

Bottom Line: However, the high variability of brain shape and size due to normal intersubject variability, normal changes occurring over the lifespan, and abnormal changes due to disease makes the ICV estimation problem challenging.To this end, an improved nonlocal label fusion scheme based on BEaST technique is proposed to increase the accuracy of the ICV estimation.The proposed method is compared with recent state-of-the-art methods and the results demonstrate an improved performance both in terms of accuracy and reproducibility while maintaining a reduced computational burden.

View Article: PubMed Central - PubMed

Affiliation: Instituto de Aplicaciones de las Tecnologías de la Información y de las Comunicaciones Avanzadas (ITACA), Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, Spain.

ABSTRACT
Automatic and accurate methods to estimate normalized regional brain volumes from MRI data are valuable tools which may help to obtain an objective diagnosis and followup of many neurological diseases. To estimate such regional brain volumes, the intracranial cavity volume (ICV) is often used for normalization. However, the high variability of brain shape and size due to normal intersubject variability, normal changes occurring over the lifespan, and abnormal changes due to disease makes the ICV estimation problem challenging. In this paper, we present a new approach to perform ICV extraction based on the use of a library of prelabeled brain images to capture the large variability of brain shapes. To this end, an improved nonlocal label fusion scheme based on BEaST technique is proposed to increase the accuracy of the ICV estimation. The proposed method is compared with recent state-of-the-art methods and the results demonstrate an improved performance both in terms of accuracy and reproducibility while maintaining a reduced computational burden.

No MeSH data available.


Related in: MedlinePlus