Cerebellar Infarction

Datar, Sudhir; Rabinstein, Alejandro A.

doi:10.1016/j.ncl.2014.07.007

Cited by 31 publications

(38 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As brainstem and cerebellar strokes occur less frequently, brainstem and cerebellar lesions were likely not included in the original training set for the automated algorithms (Chua & Kong, 1996;Datar & Rabinstein, 2014;Kase, Norrving, Levine, & Babikian, 1993;Teasell, Foley, Doherty, & Finestone, 2002). As brainstem and cerebellar strokes occur less frequently, brainstem and cerebellar lesions were likely not included in the original training set for the automated algorithms (Chua & Kong, 1996;Datar & Rabinstein, 2014;Kase, Norrving, Levine, & Babikian, 1993;Teasell, Foley, Doherty, & Finestone, 2002).…”

Section: Small Brainstem and Cerebellar Lesions Perform Worstmentioning

confidence: 99%

“…Regarding lesion location, fully automated approaches displayed significantly higher segmentation accuracy on cortical lesions than subcortical, brainstem, and cerebellar lesions, and subcortical lesions generally displayed higher segmentation accuracy than brainstem and cerebellar lesions. As brainstem and cerebellar strokes occur less frequently, brainstem and cerebellar lesions were likely not included in the original training set for the automated algorithms (Chua & Kong, 1996;Datar & Rabinstein, 2014;Kase, Norrving, Levine, & Babikian, 1993;Teasell, Foley, Doherty, & Finestone, 2002). Moreover, features implemented in the algorithms to classify lesions may not be sensitive to brainstem or cerebellar strokes.…”

Section: Small Brainstem and Cerebellar Lesions Perform Worstmentioning

confidence: 99%

See 1 more Smart Citation

A comparison of automated lesion segmentation approaches for chronic stroke T1‐weighted MRI data

Ito

Kim

Liew

2019

Human Brain Mapping

View full text Add to dashboard Cite

Accurate stroke lesion segmentation is a critical step in the neuroimaging processing pipeline for assessing the relationship between poststroke brain structure, function, and behavior. Many multimodal segmentation algorithms have been developed for acute stroke neuroimaging, yet few algorithms are effective with only a single T1‐weighted (T1w) anatomical MRI. This is a critical gap because multimodal MRI is not commonly available due to time and cost constraints in the stroke rehabilitation setting. Although several attempts to automate the segmentation of chronic lesions on single‐channel T1w MRI have been made, these approaches have not been systematically evaluated on a large dataset. We performed an exhaustive review of the literature and identified one semiautomated and three fully automated approaches for segmentation of chronic stroke lesions using T1w MRI within the last 10 years: Clusterize, automated lesion identification (ALI), Gaussian naïve Bayes lesion detection (lesionGnb), and lesion identification with neighborhood data analysis (LINDA). We evaluated each method on a large T1w stroke dataset (N = 181). LINDA was the most computationally expensive approach, but performed best across the three main evaluation metrics (median values: dice coefficient = 0.50, Hausdorff's distance = 36.34 mm, and average symmetric surface distance = 4.97 mm). lesionGnb had the highest recall/least false negatives (median = 0.80). However, across the automated methods, many lesions were either misclassified (ALI: 28, lesionGnb: 39, LINDA: 45) or not identified (ALI: 24, LINDA: 23, lesionGnb: 0). Segmentation accuracy in all automated methods were influenced by size (small: worst) and stroke territory (brainstem, cerebellum: worst) of the lesion. To facilitate reproducible science, our analysis files have been made publicly available online.

show abstract

Section: Small Brainstem and Cerebellar Lesions Perform Worstmentioning

confidence: 99%

Section: Small Brainstem and Cerebellar Lesions Perform Worstmentioning

confidence: 99%

A comparison of automated lesion segmentation approaches for chronic stroke T1‐weighted MRI data

Ito

Kim

Liew

2019

Human Brain Mapping

View full text Add to dashboard Cite

show abstract

“…Regarding lesion location, fully automated approaches displayed significantly higher segmentation accuracy on cortical lesions than subcortical, brainstem, and cerebellar lesions. As brainstem and cerebellar strokes occur less frequently, brainstem and cerebellar lesions were likely not included in the original training set for the automated algorithms (Chua & Kong, 1996;Datar & Rabinstein, 2014;Kase, Norrving, Levine, & Babikian, 1993;Teasell, Foley, Doherty, & Finestone, 2002). Moreover, features implemented in the algorithms to classify lesions (e.g., hemispheric asymmetry) may not be sensitive to brainstem or cerebellar strokes.…”

Section: Small Subcortical Brainstem or Cerebellar Lesions Perform mentioning

confidence: 99%

A comparison of automated lesion segmentation approaches for chronic stroke T1-weighted MRI data

Ito

Kim

Liew

2018

Preprint

View full text Add to dashboard Cite

Accurate stroke lesion segmentation is a critical step in the neuroimaging processing pipeline to assess the relationship between post-stroke brain structure, function, and behavior. While many multimodal segmentation algorithms have been developed for acute stroke neuroimaging, few are effective with only a single T1-weighted (T1w) anatomical MRI. This is a critical gap because most stroke rehabilitation research relies on a single T1w MRI for defining the lesion. Although several attempts to automate the segmentation of chronic lesions on single-channel T1w MRI have been made, these approaches have not been systematically evaluated on a large dataset. Here, we performed an exhaustive review of the literature and identified one semi-and three fully automated approaches for segmentation of chronic stroke lesions using T1w MRI within the last ten years: Clusterize, Automated Lesion Identification, Gaussian naïve Bayes lesion detection, and LINDA. We evaluated each method on a large T1w stroke dataset (N=181) using visual and quantitative methods. LINDA was the most computationally expensive approach, but performed best across the three main evaluation metrics (median values: Dice Coefficient=0.50, Hausdorff's Distance=36.34 mm, and Average Symmetric Surface Distance = 4.97 mm), whereas the Gaussian Bayes method had the highest recall/least false positives (median=0.80). Segmentation accuracy in all automated methods were influenced by size (small: worst) and stroke territory (brainstem, cerebellum: worst) of the lesion. To facilitate reproducible science, we have made our analysis files publicly available online at https://github.com/npnl/elsa. We hope these findings are informative to future development of T1w lesion segmentation algorithms.

show abstract

“…It may mimic benign conditions such as viral gastroenteritis or labyrinthitis, which are often seen in patients admitted to the emergency department [9]. Neurological signs associated with cerebellar infarcts include dysmetria, dysarthria, ataxia and nystagmus [10].…”

Section: Clinical Presentationmentioning

confidence: 99%

MRI of Cerebellar Infarction

2017

View full text Add to dashboard Cite

Background: MRI is the imaging modality of choice for diagnosing brain infarction. Because of few or atypical clinical symptoms and a relatively low sensitivity of CT scans, many cerebellar infarctions may be detected only with MRI. With adequate recognition of cerebellar infarction on MRI and prompt initiation or optimisation of preventive therapeutic measures, more dramatic strokes may be avoided in selected cases. Summary: We first briefly review the clinical presentation of cerebellar infarctions, followed by a short refresher on cerebellar anatomy and pathophysiological mechanisms of cerebellar infarcts. Then, we review the arterial cerebellar perfusion territories recently made visible with territorial arterial spin labeling (ASL), followed by a discussion and illustration of the MRI appearance of cerebellar infarcts in different stages. Similar to large cerebellar infarcts, recent studies investigating volumetric MRI datasets have now shown that small cerebellar infarcts occur in typical spatial patterns, knowledge of which may help in the diagnosis of even the smallest of cerebellar infarcts on MRI. Key Messages: MRI is the modality of choice for diagnosing cerebellar infarction. The posterior inferior cerebellar artery (PICA)-territories can be visualised with super-selective territorial ASL MRI. The PICA supplies at least the medial part of the posterior cerebellar surface. Anterior inferior cerebellar artery-infarcts can be mistaken for lateral PICA-infarcts. Small infarcts typically affect the cortex and often present as incidental cavities. Subacute cerebellar infarcts may be missed on imaging due to a phenomenon called “fogging.”

show abstract

Cerebellar Infarction

Cited by 31 publications

References 40 publications

A comparison of automated lesion segmentation approaches for chronic stroke T1‐weighted MRI data

A comparison of automated lesion segmentation approaches for chronic stroke T1‐weighted MRI data

A comparison of automated lesion segmentation approaches for chronic stroke T1-weighted MRI data

MRI of Cerebellar Infarction

Contact Info

Product

Resources

About