Investigating the challenges and generalizability of deep learning brain conductivity mapping
- 14 May 2020
- journal article
- research article
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 65 (13), 135001
- https://doi.org/10.1088/1361-6560/ab9356
Abstract
Purpose: To investigate deep learning electrical properties tomography (EPT) for application on different simulated and in-vivo datasets including pathologies for brain conductivity reconstructions. Methods: 3D patch-based convolutional neural networks were trained to predict conductivity maps from B1 transceive phase data. To compare the performance of DL-EPT networks on different datasets, three datasets were used throughout this work, one from simulations and two from in-vivo measurements from healthy volunteers and cancer patients, respectively. At first, networks trained on simulations were tested on all datasets with different levels of homogeneous Gaussian noise introduced in training and testing. Secondly, to investigate potential robustness towards systematical differences between simulated and measured phase maps, in-vivo data with conductivity labels from conventional EPT were used for training. Results: High quality conductivity reconstructions from networks trained on simulations with and without noise confirms the potential of deep learning for EPT. However, when this network is used for in-vivo reconstructions, measurement related artifacts affect the quality of conductivity maps. Training DL-EPT networks using conductivity labels from conventional EPT improves the quality of the results. Conclusions: Networks trained on realistic simulations yield reconstruction artifacts when applied to in-vivo data. Training with realistic phase data and conductivity labels from conventional EPT allows for reducing these artifacts.This publication has 26 references indexed in Scilit:
- Hyperthermia treatment planning for cervical cancer patients based on electrical conductivity tissue properties acquiredin vivowith EPT at 3 T MRIInternational Journal of Hyperthermia, 2016
- U-Net: Convolutional Networks for Biomedical Image SegmentationPublished by Springer Science and Business Media LLC ,2015
- Theoretical Investigation of Random Noise-Limited Signal-to-Noise Ratio in MR-Based Electrical Properties TomographyIEEE Transactions on Medical Imaging, 2015
- Deep learningNature, 2015
- CSI-EPT: A Contrast Source Inversion Approach for Improved MRI-Based Electric Properties TomographyIEEE Transactions on Medical Imaging, 2015
- Initial study on in vivo conductivity mapping of breast cancer using MRIJournal of Magnetic Resonance Imaging, 2014
- DREAM—a novel approach for robust, ultrafast, multislice B1 mappingMagnetic Resonance in Medicine, 2012
- B1 mapping by Bloch‐Siegert shiftMagnetic Resonance in Medicine, 2010
- The Virtual Family—development of surface-based anatomical models of two adults and two children for dosimetric simulationsPhysics in Medicine & Biology, 2009
- Actual flip‐angle imaging in the pulsed steady state: A method for rapid three‐dimensional mapping of the transmitted radiofrequency fieldMagnetic Resonance in Medicine, 2006