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

Comparison of one 75 years old patient with Alzheimers disease (AD) and one 68 years old healthy control (HC). (A) shows the normalized CBF distribution, (B) shows the Tscore-map computed voxel-wise. A substantial difference could be already spotted in the parietal lobe. (C) shows the likelihood of hypo-perfusion L. In the AD patient severely hypo-perfused areas in the parietal lobe and neighboring regions are visible, whereas the HC subject shows regions with artificially low perfusion in the frontotemporal lobes and in the volume boundaries. These high values are the effect of uncertainty resulting from image noise, and/or patient-specific conditions. (D) shows the posterior probability (p), which represents the disease related likelihood of hypo-perfusion resulting from the application of the prior distribution. Areas with artificially low perfusion on volume boundaries have been suppressed, while the the sensitivity to interesting hypo-perfused region in the parietal lobe is still preserved. Upper labels Rh, Cp, and Lh indicates right hemisphere, central posterior, and left hemisphere, respectively. Lateral labels Ex and In indicates external view and internal view, respectively.
© Copyright Policy
Related In: Results  -  Collection

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

Figure 2: Comparison of one 75 years old patient with Alzheimers disease (AD) and one 68 years old healthy control (HC). (A) shows the normalized CBF distribution, (B) shows the Tscore-map computed voxel-wise. A substantial difference could be already spotted in the parietal lobe. (C) shows the likelihood of hypo-perfusion L. In the AD patient severely hypo-perfused areas in the parietal lobe and neighboring regions are visible, whereas the HC subject shows regions with artificially low perfusion in the frontotemporal lobes and in the volume boundaries. These high values are the effect of uncertainty resulting from image noise, and/or patient-specific conditions. (D) shows the posterior probability (p), which represents the disease related likelihood of hypo-perfusion resulting from the application of the prior distribution. Areas with artificially low perfusion on volume boundaries have been suppressed, while the the sensitivity to interesting hypo-perfused region in the parietal lobe is still preserved. Upper labels Rh, Cp, and Lh indicates right hemisphere, central posterior, and left hemisphere, respectively. Lateral labels Ex and In indicates external view and internal view, respectively.

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)

Comparison of one 75 years old patient with Alzheimers disease (AD) and one 68 years old healthy control (HC). (A) shows the normalized CBF distribution, (B) shows the Tscore-map computed voxel-wise. A substantial difference could be already spotted in the parietal lobe. (C) shows the likelihood of hypo-perfusion L. In the AD patient severely hypo-perfused areas in the parietal lobe and neighboring regions are visible, whereas the HC subject shows regions with artificially low perfusion in the frontotemporal lobes and in the volume boundaries. These high values are the effect of uncertainty resulting from image noise, and/or patient-specific conditions. (D) shows the posterior probability (p), which represents the disease related likelihood of hypo-perfusion resulting from the application of the prior distribution. Areas with artificially low perfusion on volume boundaries have been suppressed, while the the sensitivity to interesting hypo-perfused region in the parietal lobe is still preserved. Upper labels Rh, Cp, and Lh indicates right hemisphere, central posterior, and left hemisphere, respectively. Lateral labels Ex and In indicates external view and internal view, respectively.
© Copyright Policy
Related In: Results  -  Collection

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

Figure 2: Comparison of one 75 years old patient with Alzheimers disease (AD) and one 68 years old healthy control (HC). (A) shows the normalized CBF distribution, (B) shows the Tscore-map computed voxel-wise. A substantial difference could be already spotted in the parietal lobe. (C) shows the likelihood of hypo-perfusion L. In the AD patient severely hypo-perfused areas in the parietal lobe and neighboring regions are visible, whereas the HC subject shows regions with artificially low perfusion in the frontotemporal lobes and in the volume boundaries. These high values are the effect of uncertainty resulting from image noise, and/or patient-specific conditions. (D) shows the posterior probability (p), which represents the disease related likelihood of hypo-perfusion resulting from the application of the prior distribution. Areas with artificially low perfusion on volume boundaries have been suppressed, while the the sensitivity to interesting hypo-perfused region in the parietal lobe is still preserved. Upper labels Rh, Cp, and Lh indicates right hemisphere, central posterior, and left hemisphere, respectively. Lateral labels Ex and In indicates external view and internal view, respectively.
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