caa a22 4500 52879454X 20181001030218.0 cr unu---uuuuu 180925s2018 xx s 000 0 eng 1664-2295 10.3389/fneur.2018.00717 doi (BORIS)oai:boris.unibe.ch:120084 610 ddc 30233482 pmid A Machine Learning Approach to Perfusion Imaging With Dynamic Susceptibility Contrast MR [Elektronische Daten] [Richard McKinley, Fan Hung, Roland Wiest, David S Liebeskind, Fabien Scalzo] publishedVersion openAccess eu-repo Dynamic susceptibility contrast (DSC) MR perfusion is a frequently-used technique for neurovascular imaging. The progress of a bolus of contrast agent through the tissue of the brain is imaged via a series of T2-weighted MRI scans. Clinically relevant parameters such as blood flow and Tmax can be calculated by deconvolving the contrast-time curves with the bolus shape (arterial input function). In acute stroke, for instance, these parameters may help distinguish between the likely salvageable tissue and irreversibly damaged infarct core. Deconvolution typically relies on singular value decomposition (SVD): however, studies have shown that these algorithms are very sensitive to noise and artifacts present in the image and therefore may introduce distortions that influence the estimated output parameters. In this work, we present a machine learning approach to the estimation of perfusion parameters in DSC-MRI. Various machine learning models using as input the raw MR source data were trained to reproduce the output of an FDA approved commercial implementation of the SVD deconvolution algorithm. Experiments were conducted to determine the effect of training set size, optimal patch size, and the effect of using different machine-learning models for regression. Model performance increased with training set size, but after 5,000 samples (voxels) this effect was minimal. Models inferring perfusion maps from a 5 by 5 voxel patch outperformed models able to use the information in a single voxel, but larger patches led to worse performance. Random Forest models produced had the lowest root mean squared error, with neural networks performing second best: however, a phantom study revealed that the random forest was highly susceptible to noise levels, while the neural network was more robust. The machine learning-based approach produces estimates of the perfusion parameters invariant to the noise and artifacts that commonly occur as part of MR acquisition. As a result, better robustness to noise is obtained, when evaluated against the FDA approved software on acute stroke patients and simulated phantom data. McKinley Richard joint author Hung Fan joint author Wiest Roland joint author Liebeskind David S. joint author Scalzo Fabien joint author Frontiers in neurology 9(717), p. 717 1664-2295 https://boris.unibe.ch/120084/1/fneur-09-00717.pdf text/html WWW-Backlink auf das Repository 1 info:eu-repo/semantics/article boris PeerReviewed boris peerstatus BORIS 856 40 https://boris.unibe.ch/120084/1/fneur-09-00717.pdf text/html WWW-Backlink auf das Repository BORIS 700 1- McKinley Richard joint author BORIS 700 1- Hung Fan joint author BORIS 700 1- Wiest Roland joint author BORIS 700 1- Liebeskind David S. joint author BORIS 700 1- Scalzo Fabien joint author BORIS 773 0- Frontiers in neurology 9(717), p. 717 1664-2295 BK010053 XK010053 XK010000 BORIS BORIS BORIS info:eu-repo/semantics/article openAccess