Deep-Learning-Based Accelerated and Noise-Suppressed Estimation (DANSE) of quantitative Gradient Recalled Echo (qGRE) MRI metrics associated with Human Brain Neuronal Structure and Hemodynamic Properties.

In NMR in biomedicine ; h5-index 41.0

PURPOSE : To introduce a Deep-Learning-Based Accelerated and Noise-Suppressed Estimation (DANSE) method for reconstructing quantitative maps of biological tissue cellular-specific, R2t*, and hemodynamic-specific, R2', metrics of quantitative Gradient-Recalled-Echo (qGRE) MRI.

METHODS : DANSE method adapts supervised learning paradigm to train a convolutional neural network for robust estimation of R2t* and R2' maps with significantly reduced sensitivity to noise and the adverse effects of macroscopic (B₀ ) magnetic field inhomogeneities directly from the GRE magnitude images. The R2t* and R2' maps for training were generated by means of a voxel-by-voxel fitting of a previously-developed biophysical quantitative qGRE model accounting for tissue, hemodynamic and B₀ -inhomogeneities contributions to multi-gradient-echo GRE signal using nonlinear least square (NLLS) algorithm.

RESULTS : We show that the DANSE model efficiently estimates the aforementioned qGRE maps and preserves all features of NLLS approach with significant improvements including noise-suppression and computation speed (from many hours to seconds). The noise-suppression feature of DANSE is especially prominent for data with low SNR characteristic (SNR~50-100), where DANSE-generated R2t* and R2' maps had up to three times smaller errors than that of NLLS method.

CONCLUSIONS : DANSE method enables fast reconstruction of qGRE maps with significantly reduced sensitivity to noise and magnetic field inhomogeneities. DANSE method does not require any information about field inhomogeneities during application. It exploits spatial and gradient-echo-time dependent patterns in the GRE data and previously-gained knowledge from the biophysical model, thus producing high quality qGRE maps even in the environments with high noise levels. These features along with fast computational speed can lead to broad qGRE clinical and research applications.

Kahali Sayan, Kothapalli Satya V V N, Xu Xiaojian, Kamilov Ulugbek S, Yablonskiy Dmitriy A

2022-Nov-28

BOLD, tissue microstructure, brain neuronal structure, deep learning, quantitative gradient recalled echo MRI