Feature space analysis: Effects of MRI protocols

Soltanian-Zadeh, Hamid; Peck, Donald J.

doi:10.1118/1.1414306

“…A number of algorithms exist for mapping tumor borders and functional regions within heterogeneous environment of high‐grade gliomas. Much of the segmentation research has moved toward multispectral analyses, utilizing a number of images or parametric maps offered by MRI (37–43). However, T2‐weighted signal abnormalities used independently could serve as a gauge of microscopic infiltrative disease (44), although the increased signal intensity is not specific to areas of infiltrative tumor.…”

Section: Discussionmentioning

confidence: 99%

T2 detection of tumor invasion within segmented components of glioblastoma multiforme

McMillan

¹

,

Ehtesham

²

,

Stevenson

³

et al. 2009

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Purpose: To use T2-weighted images to detect tumor invasion when comparing normal individuals to groups of gliomablastoma multiforme (GBM) patients with varying levels of CXCR4, a chemokine receptor that promotes tumor migration.Materials and Methods: T2-weighted images were acquired preoperatively in 22 treatment-naïve GBM patients. Two groups were formed based on the expression levels of CXCR4. A third group of normal volunteers was used for comparison. Each image was segmented to obtain four different clusters for tissue types identified as white matter, basal ganglia, gray matter/edema and cerebrospinal fluid (CSF)/tumor. Signal intensity histograms were formed for each cluster and compared between groups.Results: In every cluster the GBM groups displayed significantly higher standard deviations of intensity distributions when compared to normal subjects. Significant differences in skewness were found between normal subjects and GBM patients in the white matter, basal ganglia, and CSF/tumor. Further, when the two groups of GBM patients were compared the CXCR4-high group was found to have a significant shift in the median intensity values in the cluster containing gray matter and peritumoral edema. Conclusion:T2 signal intensity histograms in normal subjects differ significantly from those obtained from GBM groups, suggesting widespread dissemination of disease.

show abstract

“…In [37], the FLAIR sequence was demonstrated to be the most crucial for lesion segmentation. Despite this, it is common to analyze FLAIR via multi-modal approaches that co-register FLAIR to T1 or T2 MRI [137], [66].…”

Section: Discussionmentioning

confidence: 99%

“…This multiparametric approach prolongs scan times, increases acquisition costs, and can introduce registration errors across sequences [37], [137], [66] for WML segmentation approaches. Since FLAIR is routinely acquired clinically and highlights vascular disease with high sensitivity, there is benefit in developing methods that operate on this single sequence.…”

Section: Discussionmentioning

confidence: 99%

Neurological Symmetry Analysis Applications in Flair MRI

Gibicar

2024

Preprint

0

View full text Add to dashboard Cite

<p>Neurological diseases are one of the leading causes of disability in the global population. Fluid Attenuated Inversion Recovery (FLAIR) MRI is an important imaging modality for diagnosing neurological diseases such as Alzheimer's disease (AD) and cerebrovascular disease (CVD). Analyzing characteristics of neurological diseases using FLAIR MRI provide useful neuroimaging biomarkers for diagnosis, disease monitoring/progression, and treatment planning. There is growing evidence that these neurological diseases may affect cerebral symmetry. Thus, this thesis presents the development of automated segmentation tools to facilitate clinical symmetry analysis of neuroimaging biomarkers in AD and CVD cohorts. To do this, a fully unsupervised midline estimation technique was proposed, validated against existing works, and yielded the lowest average error across all metrics: Hausdorff distance (HD) = 0.32 ¬± 0.23, mean absolute difference (MAD) = 1.10 ¬± 0.38 mm and volume difference = 7.52 ¬± 5.40 and 5.35 ¬± 3.97 ml, for left and right hemispheres, respectively. Next, white matter lesion segmentation algorithms were explored and evaluated to later develop biomarkers for CVD subjects. Of all methods explored, the SC U-Net architecture was found to be the best algorithm for WML segmentation in terms of highest Dice similarity coefficient (DSC) over most dimensions (mean DSC = 0.71 over all volumes). These automated segmentation tools enabled downstream clinical symmetry analysis. Using the midline estimation tool, hemispheric asymmetry texture biomarkers of the normal-appearing brain matter (NABM) were analyzed in a dementia cohort, and textural asymmetry was found to increase with the progression of AD. Symmetry analysis was also performed in a CVD cohort, using both the midline and WML algorithms. Significant associations between WML asymmetry and carotid disease biomarkers were found. This work demonstrates that automated algorithms developed for FLAIR MRI can be used to measure asymmetrical properties of neurological diseases, that may aid in the detection of disease, treatment planning and clinical outcomes.</p>

show abstract

“…In [37], the FLAIR sequence was demonstrated to be the most crucial for lesion segmentation. Despite this, it is common to analyze FLAIR via multi-modal approaches that co-register FLAIR to T1 or T2 MRI [137], [66].…”

Section: Discussionmentioning

confidence: 99%

“…This multiparametric approach prolongs scan times, increases acquisition costs, and can introduce registration errors across sequences [37], [137], [66] for WML segmentation approaches. Since FLAIR is routinely acquired clinically and highlights vascular disease with high sensitivity, there is benefit in developing methods that operate on this single sequence.…”

Section: Discussionmentioning

confidence: 99%

Neurological Symmetry Analysis Applications in Flair MRI

Gibicar

2024

Preprint

0

View full text Add to dashboard Cite

<p>Neurological diseases are one of the leading causes of disability in the global population. Fluid Attenuated Inversion Recovery (FLAIR) MRI is an important imaging modality for diagnosing neurological diseases such as Alzheimer's disease (AD) and cerebrovascular disease (CVD). Analyzing characteristics of neurological diseases using FLAIR MRI provide useful neuroimaging biomarkers for diagnosis, disease monitoring/progression, and treatment planning. There is growing evidence that these neurological diseases may affect cerebral symmetry. Thus, this thesis presents the development of automated segmentation tools to facilitate clinical symmetry analysis of neuroimaging biomarkers in AD and CVD cohorts. To do this, a fully unsupervised midline estimation technique was proposed, validated against existing works, and yielded the lowest average error across all metrics: Hausdorff distance (HD) = 0.32 ¬± 0.23, mean absolute difference (MAD) = 1.10 ¬± 0.38 mm and volume difference = 7.52 ¬± 5.40 and 5.35 ¬± 3.97 ml, for left and right hemispheres, respectively. Next, white matter lesion segmentation algorithms were explored and evaluated to later develop biomarkers for CVD subjects. Of all methods explored, the SC U-Net architecture was found to be the best algorithm for WML segmentation in terms of highest Dice similarity coefficient (DSC) over most dimensions (mean DSC = 0.71 over all volumes). These automated segmentation tools enabled downstream clinical symmetry analysis. Using the midline estimation tool, hemispheric asymmetry texture biomarkers of the normal-appearing brain matter (NABM) were analyzed in a dementia cohort, and textural asymmetry was found to increase with the progression of AD. Symmetry analysis was also performed in a CVD cohort, using both the midline and WML algorithms. Significant associations between WML asymmetry and carotid disease biomarkers were found. This work demonstrates that automated algorithms developed for FLAIR MRI can be used to measure asymmetrical properties of neurological diseases, that may aid in the detection of disease, treatment planning and clinical outcomes.</p>

show abstract

Feature space analysis: Effects of MRI protocols

Cited by 11 publications

References 21 publications

T2 detection of tumor invasion within segmented components of glioblastoma multiforme

T2 detection of tumor invasion within segmented components of glioblastoma multiforme

Neurological Symmetry Analysis Applications in Flair MRI

Neurological Symmetry Analysis Applications in Flair MRI

Contact Info

Product

Resources

About