Limits...
Diagnostic Potential of Pulsed Arterial Spin Labeling in Alzheimer's Disease.

Trebeschi S, Riederer I, Preibisch C, Bohn KP, Förster S, Alexopoulos P, Zimmer C, Kirschke JS, Valentinitsch A - Front Neurosci (2016)

Bottom Line: The discriminant analysis is carried out to maximize the accuracy of the classification.The algorithm has been trained on a dataset of 81 subjects and achieved a sensitivity of 0.750 and a specificity of 0.875.Moreover, in accordance with the current pathological knowledge, the parietal lobe, and limbic system are shown to be the main discriminant factors.

View Article: PubMed Central - PubMed

Affiliation: Department of Neuroradiology, Klinikum rechts der Isar, Technische Universität München Munich, Germany.

ABSTRACT
Alzheimers disease (AD) is the most common cause of dementia. Although the underlying pathology is still not completely understood, several diagnostic methods are available. Frequently, the most accurate methods are also the most invasive. The present work investigates the diagnostic potential of Pulsed Arterial Spin Labeling (PASL) for AD: a non-invasive, MRI-based technique for the quantification of regional cerebral blood flow (rCBF). In particular, we propose a pilot computer aided diagnostic (CAD) procedure able to discriminate between healthy and diseased subjects, and at the same time, providing visual informative results. This method encompasses the creation of a healthy model, the computation of a voxel-wise likelihood function as comparison between the healthy model and the subject under examination, and the correction of the likelihood function via prior distributions. The discriminant analysis is carried out to maximize the accuracy of the classification. The algorithm has been trained on a dataset of 81 subjects and achieved a sensitivity of 0.750 and a specificity of 0.875. Moreover, in accordance with the current pathological knowledge, the parietal lobe, and limbic system are shown to be the main discriminant factors.

No MeSH data available.


Related in: MedlinePlus

Model of the cerebral blood perfusion based on HC subjects: (A) shows the mean and (B) the standard deviation of the healthy group. High blood perfusion is visible in the parietal lobe and in the cuneus. We included for visual comparison (C), which shows the mean perfusion of the AD cohort. (D) visualization of the prior distribution using principal component analysis to depict the predictive power of certain cerebral regions, where the parietal lobe showed the highest predictive power.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4835490&req=5

Figure 3: Model of the cerebral blood perfusion based on HC subjects: (A) shows the mean and (B) the standard deviation of the healthy group. High blood perfusion is visible in the parietal lobe and in the cuneus. We included for visual comparison (C), which shows the mean perfusion of the AD cohort. (D) visualization of the prior distribution using principal component analysis to depict the predictive power of certain cerebral regions, where the parietal lobe showed the highest predictive power.

Mentions: This observation is also confirmed by the prior distribution built upon a regression model (Figure 2D) where the parietal lobe and the surrounding regions evidenced a higher importance for the prediction of AD. Figure 2 shows a comparison between two randomly selected HC and AD subjects. Figures 2A,B represent the CBF and the T-score, respectively. Figure 2C shows the likelihood L using voxel-wise comparison. The healthy individual showed low perfusion in the frontotemporal lobe, which results from poor image quality due to susceptibility artifacts in these areas. We could reduce this effect of uncertainty in the likelihood images L by adding a prior distribution (Figure 3D). In the posterior images (P) the areas with low perfusion in healthy subjects have been suppressed, whereas regions of hypo-perfusion in the parietal lobes of AD subject were unaffected.


Diagnostic Potential of Pulsed Arterial Spin Labeling in Alzheimer's Disease.

Trebeschi S, Riederer I, Preibisch C, Bohn KP, Förster S, Alexopoulos P, Zimmer C, Kirschke JS, Valentinitsch A - Front Neurosci (2016)

Model of the cerebral blood perfusion based on HC subjects: (A) shows the mean and (B) the standard deviation of the healthy group. High blood perfusion is visible in the parietal lobe and in the cuneus. We included for visual comparison (C), which shows the mean perfusion of the AD cohort. (D) visualization of the prior distribution using principal component analysis to depict the predictive power of certain cerebral regions, where the parietal lobe showed the highest predictive power.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4835490&req=5

Figure 3: Model of the cerebral blood perfusion based on HC subjects: (A) shows the mean and (B) the standard deviation of the healthy group. High blood perfusion is visible in the parietal lobe and in the cuneus. We included for visual comparison (C), which shows the mean perfusion of the AD cohort. (D) visualization of the prior distribution using principal component analysis to depict the predictive power of certain cerebral regions, where the parietal lobe showed the highest predictive power.
Mentions: This observation is also confirmed by the prior distribution built upon a regression model (Figure 2D) where the parietal lobe and the surrounding regions evidenced a higher importance for the prediction of AD. Figure 2 shows a comparison between two randomly selected HC and AD subjects. Figures 2A,B represent the CBF and the T-score, respectively. Figure 2C shows the likelihood L using voxel-wise comparison. The healthy individual showed low perfusion in the frontotemporal lobe, which results from poor image quality due to susceptibility artifacts in these areas. We could reduce this effect of uncertainty in the likelihood images L by adding a prior distribution (Figure 3D). In the posterior images (P) the areas with low perfusion in healthy subjects have been suppressed, whereas regions of hypo-perfusion in the parietal lobes of AD subject were unaffected.

Bottom Line: The discriminant analysis is carried out to maximize the accuracy of the classification.The algorithm has been trained on a dataset of 81 subjects and achieved a sensitivity of 0.750 and a specificity of 0.875.Moreover, in accordance with the current pathological knowledge, the parietal lobe, and limbic system are shown to be the main discriminant factors.

View Article: PubMed Central - PubMed

Affiliation: Department of Neuroradiology, Klinikum rechts der Isar, Technische Universität München Munich, Germany.

ABSTRACT
Alzheimers disease (AD) is the most common cause of dementia. Although the underlying pathology is still not completely understood, several diagnostic methods are available. Frequently, the most accurate methods are also the most invasive. The present work investigates the diagnostic potential of Pulsed Arterial Spin Labeling (PASL) for AD: a non-invasive, MRI-based technique for the quantification of regional cerebral blood flow (rCBF). In particular, we propose a pilot computer aided diagnostic (CAD) procedure able to discriminate between healthy and diseased subjects, and at the same time, providing visual informative results. This method encompasses the creation of a healthy model, the computation of a voxel-wise likelihood function as comparison between the healthy model and the subject under examination, and the correction of the likelihood function via prior distributions. The discriminant analysis is carried out to maximize the accuracy of the classification. The algorithm has been trained on a dataset of 81 subjects and achieved a sensitivity of 0.750 and a specificity of 0.875. Moreover, in accordance with the current pathological knowledge, the parietal lobe, and limbic system are shown to be the main discriminant factors.

No MeSH data available.


Related in: MedlinePlus