Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events
Le, Elizabeth P. V.
Tarkin, Jason M.
Evans, Nicholas R.
Chowdhury, Mohammed M.
Coughlin, Patrick A.
Gallagher, Ferdia A.
Weir-McCall, Jonathan R.
Gilbert, Fiona J.
Warburton, Elizabeth A.
Rudd, James H. F.
Nature Publishing Group UK
MetadataShow full item record
Le, E. P. V., Rundo, L., Tarkin, J. M., Evans, N. R., Chowdhury, M. M., Coughlin, P. A., Pavey, H., et al. (2021). Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events. Scientific Reports, 11 (1) https://doi.org/10.1038/s41598-021-82760-w
Funder: School of Clinical Medicine, University of Cambridge; doi: http://dx.doi.org/10.13039/501100007552
Funder: Frank Edward Elmore Fund
Funder: National Institute for Health Research (NIHR) Imperial Biomedical Research Centre
Funder: British Heart Foundation Cambridge Centre of Research Excellence
Funder: Royal College of Surgeons of England; doi: http://dx.doi.org/10.13039/501100000297
Funder: Cancer Research UK; doi: http://dx.doi.org/10.13039/501100000289
Funder: AstraZeneca Oncology R
Funder: National Institute for Health Research; doi: http://dx.doi.org/10.13039/501100000272
Funder: Leverhulme Trust; doi: http://dx.doi.org/10.13039/501100000275
Funder: Cantab Capital Institute for the Mathematics of Information
Funder: Alan Turing Institute; doi: http://dx.doi.org/10.13039/100012338
Funder: NIHR Cambridge Biomedical Research Centre
Funder: Higher Education Funding Council for England; doi: http://dx.doi.org/10.13039/501100000384
Abstract: Radiomics, quantitative feature extraction from radiological images, can improve disease diagnosis and prognostication. However, radiomic features are susceptible to image acquisition and segmentation variability. Ideally, only features robust to these variations would be incorporated into predictive models, for good generalisability. We extracted 93 radiomic features from carotid artery computed tomography angiograms of 41 patients with cerebrovascular events. We tested feature robustness to region-of-interest perturbations, image pre-processing settings and quantisation methods using both single- and multi-slice approaches. We assessed the ability of the most robust features to identify culprit and non-culprit arteries using several machine learning algorithms and report the average area under the curve (AUC) from five-fold cross validation. Multi-slice features were superior to single for producing robust radiomic features (67 vs. 61). The optimal image quantisation method used bin widths of 25 or 30. Incorporating our top 10 non-redundant robust radiomics features into ElasticNet achieved an AUC of 0.73 and accuracy of 69% (compared to carotid calcification alone [AUC: 0.44, accuracy: 46%]). Our results provide key information for introducing carotid CT radiomics into clinical practice. If validated prospectively, our robust carotid radiomic set could improve stroke prediction and target therapies to those at highest risk.
Article, /692/308, /692/700/1421, /692/4019/592/75/593/2100, /692/699/75/593/1353, article
Medical Research Council (1966157)
The Mark Foundation for Cancer Research and Cancer Research UK (CRUK) Cambridge Centre (C9685/A25177, C9685/A25177)
Wellcome Trust (211100/Z/18/Z)
The Dunhill Medical Trust (RTF44/0114)
British Heart Foundation (FS/16/29/31957)
EPSRC (EP/S026045/1 and EP/T003553/1, EP/N014588/1)
Wellcome Innovator Award (RG98755)
Horizon 2020 (No. 777826 NoMADS and No. 691070 CHiPS)
External DOI: https://doi.org/10.1038/s41598-021-82760-w
This record's URL: https://www.repository.cam.ac.uk/handle/1810/317456
Attribution 4.0 International (CC BY 4.0)
Licence URL: https://creativecommons.org/licenses/by/4.0/