4.5 Article

Automated Detection of Candidate Subjects With Cerebral Microbleeds Using Machine Learning

Journal

FRONTIERS IN NEUROINFORMATICS
Volume 15, Issue -, Pages -

Publisher

FRONTIERS MEDIA SA
DOI: 10.3389/fninf.2021.777828

Keywords

cerebral microbleeds; machine learning; susceptibility weighted image (SWI); UK Biobank; subject-level detection; T2*-weighted MRI; structural MRI

Funding

  1. Wellcome Trust
  2. Engineering and Physical Sciences Research Council (EPSRC), Medical Research Council (MRC) [EP/L016052/1]
  3. NIHR Nottingham Biomedical Research Centre and Wellcome Centre for Integrative Neuroimaging [215573/Z/19/Z]
  4. National Institute for Health Research (NIHR) [203141/Z/16/Z]
  5. Wolfson Foundation
  6. British Heart Foundation [PG/14/96/31262]
  7. European Union's Horizon 2020 [666881]
  8. UK National Institute for Health Research Health Technology Assessment programme [11_129_109]
  9. Wellcome Centre for Integrative Neuroimaging
  10. Ministry of Education (MIUR)
  11. Department of Biomedical, Metabolic and Neural Sciences, University of Modena and Reggio Emilia
  12. Senior Research Fellowship from the Wellcome Trust [202788/Z/16/Z]
  13. NIHR Oxford Biomedical Research Centre (BRC)
  14. Oxford Parkinson's Disease Centre (Parkinson's UK Monument Discovery Award, [J-1403]
  15. MRC Dementias Platform UK [MR/L023784/2]
  16. National Institute for Health Research (NIHR) Oxford Health Biomedical Research Centre (BRC)
  17. [36509]
  18. [43822]
  19. [[203139/Z/16/Z]]

Ask authors/readers for more resources

The study aimed to detect CMB candidate subjects from the UK Biobank dataset using a machine learning-based, computationally light pipeline. Evaluation results showed that our method achieved subject-level detection accuracy of over 80% on all datasets, and demonstrated good generalisability across datasets.
Cerebral microbleeds (CMBs) appear as small, circular, well defined hypointense lesions of a few mm in size on T2*-weighted gradient recalled echo (T2*-GRE) images and appear enhanced on susceptibility weighted images (SWI). Due to their small size, contrast variations and other mimics (e.g., blood vessels), CMBs are highly challenging to detect automatically. In large datasets (e.g., the UK Biobank dataset), exhaustively labelling CMBs manually is difficult and time consuming. Hence it would be useful to preselect candidate CMB subjects in order to focus on those for manual labelling, which is essential for training and testing automated CMB detection tools on these datasets. In this work, we aim to detect CMB candidate subjects from a larger dataset, UK Biobank, using a machine learning-based, computationally light pipeline. For our evaluation, we used 3 different datasets, with different intensity characteristics, acquired with different scanners. They include the UK Biobank dataset and two clinical datasets with different pathological conditions. We developed and evaluated our pipelines on different types of images, consisting of SWI or GRE images. We also used the UK Biobank dataset to compare our approach with alternative CMB preselection methods using non-imaging factors and/or imaging data. Finally, we evaluated the pipeline's generalisability across datasets. Our method provided subject-level detection accuracy > 80% on all the datasets (within-dataset results), and showed good generalisability across datasets, providing a consistent accuracy of over 80%, even when evaluated across different modalities.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.5
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available