Ensemble of Convolutional Neural Networks Improves Automated Segmentation of Acute Ischemic Lesions Using Multiparametric Diffusion-Weighted MRI.

Stefan Winzeck S J T Mocking R Bezerra Mark J R J Bouts Elissa C McIntosh I Diwan P Garg A Chutinet William T Kimberly William A Copen Pamela W Schaefer H Ay Aneesh Bhim Singhal Konstantinos Kamnitsas Ben Glocker A Gregory Sorensen Ona Wu

Published in: AJNR. American journal of neuroradiology (2019)

Acute infarcts are more accurately segmented using ensembles of convolutional neural networks trained with multiparametric maps than by using a single model trained with a solo map. Automated lesion segmentation has high agreement with manual techniques for identifying patients with large lesion volumes.

Keyphrases

convolutional neural network
deep learning
diffusion weighted
contrast enhanced
liver failure
respiratory failure
machine learning
magnetic resonance imaging
aortic dissection
resistance training
drug induced
high throughput
computed tomography
intensive care unit
single cell