Contrast-Based Fully Automatic Segmentation of White Matter Hyperintensities: Method and Validation

Objective: Reliable quantification of white matter hyperintensities (WHMs) resulting from cerebral small vessel diseases (CSVD) is essential for understanding their clinical impact. We aim to develop and clinically validate a deep learning system for automatic segmentation of CSVD-WMH from fluid-attenuated inversion recovery (FLAIR) imaging using large multicenter data.Method: A FLAIR imaging dataset of 1,156 patients diagnosed with CSVD associated WMH (median age, 54 years; 653 males) obtained between September 2018 and September 2019 from Beijing Tiantan Hospital was retrospectively analyzed in this study. Locations of CSVD-WMH on the FLAIR scans were manually marked by two experienced neurologists. Using the manually labeled data of 996 patients (development set), a U-shaped novel 2D convolutional neural network (CNN) architecture was trained for automatic segmentation of CSVD-WMH. The segmentation performance of the network was evaluated with per pixel and lesion level dice scores using an independent internal test set (n = 160) and a multi-center external test set (n = 90, three medical centers). The clinical suitability of the segmentation results, classified as acceptable, acceptable with minor revision, acceptable with major revision, and not acceptable, was analyzed by three independent neuroradiologists. The inter-neuroradiologists agreement rate was assessed by the Kendall-W test.Results: On the internal and external test sets, the proposed CNN architecture achieved per pixel and lesion level dice scores of 0.72 (external test set), and they were significantly better than the state-of-the-art deep learning architectures proposed for WMH segmentation. In the clinical evaluation, neuroradiologists observed the segmentation results for 95% of the patients were acceptable or acceptable with a minor revision.Conclusions: A deep learning system can be used for automated, objective, and clinically meaningful segmentation of CSVD-WMH with high accuracy.

Download Full-text

Investigating the Contribution of White Matter Hyperintensities and Cortical Thickness to Empathy in Neurodegenerative and Cerebrovascular Diseases

10.1101/2021.08.01.454640 ◽

2021 ◽

Author(s):

Miracle Ozzoude ◽

Brenda Varriano ◽

Derek Beaton ◽

Joel Ramirez ◽

Melissa F Holmes ◽

...

Keyword(s):

White Matter ◽

Cortical Thickness ◽

Brain Atrophy ◽

White Matter Hyperintensities ◽

Automatic Segmentation ◽

Cerebrovascular Diseases ◽

Cortical Atrophy ◽

Reactivity Index ◽

Empathy Deficits ◽

Longitudinal Effects

Introduction: Change in empathy is an increasingly recognised symptom of neurodegenerative diseases and contributes to caregiver burden and patient distress. Empathy impairment has been associated with brain atrophy but its relationship to white matter hyperintensities (WMH) is unknown. We aimed to investigate the relationships amongst WMH, brain atrophy, and empathy deficits in neurodegenerative and cerebrovascular diseases. Methods: 513 participants with Alzheimers Disease/Mild Cognitive Impairment, Amyotrophic Lateral Sclerosis, Frontotemporal Dementia (FTD), Parkinsons Disease, or Cerebrovascular Disease (CVD) were included. Empathy was assessed using the Interpersonal Reactivity Index. WMH were measured using a semi-automatic segmentation and FreeSurfer was used to measure cortical thickness. Results: A heterogeneous pattern of cortical thinning was found between groups, with FTD showing thinning in frontotemporal regions and CVD in left superior parietal, left insula, and left postcentral. Results from both univariate and multivariate analyses revealed that several variables were associated with empathy, particularly cortical thickness in the fronto-insulo-temporal and cingulate regions, sex(female), global cognition, and right parietal and occipital WMH. Conclusions: Our results suggest that cortical atrophy and WMH may be associated with empathy deficits in neurodegenerative and cerebrovascular diseases. Future work should consider investigating the longitudinal effects of WMH and atrophy on empathy deficits in neurodegenerative and cerebrovascular diseases.

Download Full-text

Automatic segmentation of white matter hyperintensities by an extended FitzHugh & nagumo reaction diffusion model

Journal of Magnetic Resonance Imaging ◽

10.1002/jmri.23836 ◽

2012 ◽

Vol 37 (2) ◽

pp. 343-350 ◽

Cited By ~ 6

Author(s):

Shuangxi Ji ◽

Changqing Ye ◽

Fan Li ◽

Wei Sun ◽

Jue Zhang ◽

...

Keyword(s):

White Matter ◽

Diffusion Model ◽

White Matter Hyperintensities ◽

Automatic Segmentation ◽

Reaction Diffusion ◽

Reaction Diffusion Model

Download Full-text

Automatic Segmentation of White Matter Hyperintensities from Brain Magnetic Resonance Images in the Era of Deep Learning and Big Data – A Systematic Review

10.20944/preprints202009.0210.v2 ◽

2020 ◽

Author(s):

Ramya Balakrishnan ◽

Maria Valdes Hernandez ◽

Andrew Farrall

Keyword(s):

Systematic Review ◽

Deep Learning ◽

White Matter ◽

Clinical Research ◽

White Matter Hyperintensities ◽

Segmentation Method ◽

Unsupervised Segmentation ◽

Vascular Origin ◽

Automatic Methods ◽

Fully Automatic

Background: White matter hyperintensities (WMH), of presumed vascular origin, are visible and quantifiable neuroradiological markers of brain parenchymal change. These changes may range from damage secondary to inflammation and other neurological conditions, through to healthy ageing. Fully automatic WMH quantification methods are promising, but still, traditional semi-automatic methods seem to be preferred in clinical research. We systematically reviewed the literature for fully automatic methods developed in the last five years, to assess what are considered state-of-the-art techniques, as well as trends in the analysis of WMH of presumed vascular origin. Method: We registered the systematic review protocol with the International Prospective Register of Systematic Reviews (PROSPERO), registration number - CRD42019132200. We conducted the search for fully automatic methods developed from 2015 to July 2020 on Medline, Science direct, IEE Explore, and Web of Science. We assessed risk of bias and applicability of the studies using QUADAS 2. Results: The search yielded 2327 papers after removing 104 duplicates. After screening titles, abstracts and full text, 37 were selected for detailed analysis. Of these, 16 proposed a supervised segmentation method, 10 proposed an unsupervised segmentation method, and 11 proposed a deep learning segmentation method. Average DSC values ranged from 0.538 to 0.91, being the highest value obtained from an unsupervised segmentation method. Only four studies validated their method in longitudinal samples, and eight performed an additional validation using clinical parameters. Only 8/37 studies made available their method in public repositories. Conclusions: We found no evidence that favours deep learning methods over the more established k-NN, linear regression and unsupervised methods in this task. Data and code availability, bias in study design and ground truth generation influence the wider validation and applicability of these methods in clinical research.

Download Full-text

Association of white matter hyperintensities volume and cognitive function: fully automatic volumetric measurement

RöFo - Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden Verfahren ◽

10.1055/s-0029-1221753 ◽

2009 ◽

Vol 181 (S 01) ◽

Author(s):

K Lutz ◽

R Henze ◽

J Schröder ◽

A Hanekamp ◽

M Essig

Keyword(s):

Cognitive Function ◽

White Matter ◽

White Matter Hyperintensities ◽

Volumetric Measurement ◽

Fully Automatic

Download Full-text

Automatic segmentation of white matter hyperintensities from brain magnetic resonance images in the era of deep learning and big data – A systematic review

Computerized Medical Imaging and Graphics ◽

10.1016/j.compmedimag.2021.101867 ◽

2021 ◽

Vol 88 ◽

pp. 101867

Author(s):

Ramya Balakrishnan ◽

Maria del C. Valdés Hernández ◽

Andrew J. Farrall

Keyword(s):

Systematic Review ◽

Big Data ◽

Deep Learning ◽

White Matter ◽

Magnetic Resonance ◽

White Matter Hyperintensities ◽

Automatic Segmentation ◽

Magnetic Resonance Images

Download Full-text

Automatic Segmentation and Quantitative Analysis of White Matter Hyperintensities on FLAIR Images Using Trimmed-Likelihood Estimator

Academic Radiology ◽

10.1016/j.acra.2014.07.001 ◽

2014 ◽

Vol 21 (12) ◽

pp. 1512-1523 ◽

Cited By ~ 2

Author(s):

Rui Wang ◽

Chao Li ◽

Jie Wang ◽

Xiaoer Wei ◽

Yuehua Li ◽

...

Keyword(s):

Quantitative Analysis ◽

White Matter ◽

White Matter Hyperintensities ◽

Automatic Segmentation ◽

Likelihood Estimator

Download Full-text

Automatic Segmentation of White Matter Hyperintensities from Brain Magnetic Resonance Images in the Era of Deep Learning and Big Data – A Systematic Review

10.20944/preprints202009.0210.v1 ◽

2020 ◽

Author(s):

Ramya Balakrishnan ◽

Maria Valdes Hernandez ◽

Andrew Farrall

Keyword(s):

Systematic Review ◽

Deep Learning ◽

White Matter ◽

Clinical Research ◽

White Matter Hyperintensities ◽

Magnetic Resonance Images ◽

Segmentation Method ◽

Vascular Origin ◽

Automatic Methods ◽

Fully Automatic

Background: White matter hyperintensities (WMH), of presumed vascular origin, are visible and quantifiable neuroradiological markers of brain parenchymal change. These changes may range from damage secondary to inflammation and other neurological conditions, through to healthy ageing. Fully automatic WMH quantification methods are promising, but still, traditional semi-automatic methods seem to be preferred in clinical research. We systematically reviewed the literature for fully automatic methods developed in the last five years, to assess what are considered state-of-the-art techniques, as well as trends in the analysis of WMH of presumed vascular origin. Method: We registered the systematic review protocol with the International Prospective Register of Systematic Reviews (PROSPERO), registration number - CRD42019132200. We conducted the search for fully automatic methods developed from 2015 to July 2020 on Medline, Science direct, IEE Explore, and Web of Science. We assessed risk of bias and applicability of the studies using QUADAS 2. Results: The search yielded 2327 papers after removing 104 duplicates. After screening titles, abstracts and full text, 37 were selected for detailed analysis. Of these, 16 proposed a supervised segmentation method, 10 proposed an unsupervised segmentation method, and 11 proposed a deep learning segmentation method. Average DSC values ranged from 0.538 to 0.93, being the highest value obtained from a deep learning segmentation method. Only four studies validated their method in longitudinal samples, and eight performed an additional validation using clinical parameters. Only 8/37 studies made available their method in public repositories. Conclusions: Although deep learning methods reported highly accurate results, we found no evidence that favours them over the more established k-NN, linear regression and unsupervised methods in this task. Data and code availability, bias in study design and ground truth generation influence the wider validation and applicability of these methods in clinical research.

Download Full-text