Machine Learning for Large-Scale Quality Control of 3D Shape Models in Neuroimaging.

Details

Serval ID
serval:BIB_2D96CDB2F373
Type
Article: article from journal or magazin.
Collection
Publications
Institution
Title
Machine Learning for Large-Scale Quality Control of 3D Shape Models in Neuroimaging.
Journal
Machine learning in medical imaging. MLMI
Author(s)
Petrov D., Gutman B.A., Yu S.J., van Erp TGM, Turner J.A., Schmaal L., Veltman D., Wang L., Alpert K., Isaev D., Zavaliangos-Petropulu A., Ching CRK, Calhoun V., Glahn D., Satterthwaite T.D., Andreasen O.A., Borgwardt S., Howells F., Groenewold N., Voineskos A., Radua J., Potkin S.G., Crespo-Facorro B., Tordesillas-Gutiérrez D., Shen L., Lebedeva I., Spalletta G., Donohoe G., Kochunov P., Rosa PGP, James A., Dannlowski U., Baune B.T., Aleman A., Gotlib I.H., Walter H., Walter M., Soares J.C., Ehrlich S., Gur R.C., Doan N.T., Agartz I., Westlye L.T., Harrisberger F., Riecher-Rössler A., Uhlmann A., Stein D.J., Dickie E.W., Pomarol-Clotet E., Fuentes-Claramonte P., Canales-Rodríguez E.J., Salvador R., Huang A.J., Roiz-Santiañez R., Cong S., Tomyshev A., Piras F., Vecchio D., Banaj N., Ciullo V., Hong E., Busatto G., Zanetti M.V., Serpa M.H., Cervenka S., Kelly S., Grotegerd D., Sacchet M.D., Veer I.M., Li M., Wu M.J., Irungu B., Walton E., Thompson P.M.
Publication state
Published
Issued date
09/2017
Peer-reviewed
Oui
Volume
10541
Pages
371-378
Language
english
Notes
Publication types: Journal Article
Publication Status: ppublish
Abstract
As very large studies of complex neuroimaging phenotypes become more common, human quality assessment of MRI-derived data remains one of the last major bottlenecks. Few attempts have so far been made to address this issue with machine learning. In this work, we optimize predictive models of quality for meshes representing deep brain structure shapes. We use standard vertex-wise and global shape features computed homologously across 19 cohorts and over 7500 human-rated subjects, training kernelized Support Vector Machine and Gradient Boosted Decision Trees classifiers to detect meshes of failing quality. Our models generalize across datasets and diseases, reducing human workload by 30-70%, or equivalently hundreds of human rater hours for datasets of comparable size, with recall rates approaching inter-rater reliability.
Keywords
machine learning, quality control, shape analysis
Pubmed
Create date
31/07/2018 10:56
Last modification date
20/08/2019 13:12
Usage data