Deep learning-based Accelerated and Noise-Suppressed Estimation (DANSE) of quantitative Gradient-Recalled Echo (qGRE) magnetic resonance imaging metrics associated with human brain neuronal structure and hemodynamic properties.

The purpose of the current study was 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. The DANSE method adapts a 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 gradient-recalled echo (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 multigradient-echo GRE signal using a nonlinear least squares (NLLS) algorithm. We show that the DANSE model efficiently estimates the aforementioned qGRE maps and preserves all the features of the 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 signal-to-noise ratio (SNR ~ 50-100), where DANSE-generated R2t* and R2' maps had up to three times smaller errors than that of the NLLS method. The DANSE method enables fast reconstruction of qGRE maps with significantly reduced sensitivity to noise and magnetic field inhomogeneities. The 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 environments with high noise levels. These features along with fast computational speed can lead to broad qGRE clinical and research applications.

Deep Learning and Transfer Learning for Malaria Detection.

Infectious disease malaria is a devastating infectious disease that claims the lives of more than 500,000 people worldwide every year. Most of these deaths occur as a result of a delayed or incorrect diagnosis. At the moment, the manual microscope is considered to be the most effective equipment for diagnosing malaria. It is, on the other hand, time-consuming and prone to human error. Because it is such a serious global health issue, it is important that the evaluation process be automated. The objective of this article is to advocate for the automation of the diagnosis process in order to eliminate the need for human intervention in the process. Convolutional neural networks (CNNs) and other deep-learning technologies, such as image processing, are being utilized to evaluate parasitemia in microscopic blood slides in order to enhance diagnostic accuracy. The approach is based on the intensity characteristics of Plasmodium parasites and erythrocytes, which are both known to be variable. Images of infected and noninfected erythrocytes are gathered and fed into the CNN models ResNet50, ResNet34, VGG-16, and VGG-19, which are all trained on the same dataset. The techniques of transfer learning and fine-tuning are employed, and the outcomes are contrasted. The VGG-19 model obtained the best overall performance given the parameters and dataset that were evaluated.

Learning-based motion artifact removal networks for quantitative R2∗ mapping.

PURPOSE: To introduce two novel learning-based motion artifact removal networks (LEARN) for the estimation of quantitative motion- and B0 -inhomogeneity-corrected R2∗ maps from motion-corrupted multi-Gradient-Recalled Echo (mGRE) MRI data. METHODS: We train two convolutional neural networks (CNNs) to correct motion artifacts for high-quality estimation of quantitative B0 -inhomogeneity-corrected R2∗ maps from mGRE sequences. The first CNN, LEARN-IMG, performs motion correction on complex mGRE images, to enable the subsequent computation of high-quality motion-free quantitative R2∗ (and any other mGRE-enabled) maps using the standard voxel-wise analysis or machine learning-based analysis. The second CNN, LEARN-BIO, is trained to directly generate motion- and B0 -inhomogeneity-corrected quantitative R2∗ maps from motion-corrupted magnitude-only mGRE images by taking advantage of the biophysical model describing the mGRE signal decay. RESULTS: We show that both CNNs trained on synthetic MR images are capable of suppressing motion artifacts while preserving details in the predicted quantitative R2∗ maps. Significant reduction of motion artifacts on experimental in vivo motion-corrupted data has also been achieved by using our trained models. CONCLUSION: Both LEARN-IMG and LEARN-BIO can enable the computation of high-quality motion- and B0 -inhomogeneity-corrected R2∗ maps. LEARN-IMG performs motion correction on mGRE images and relies on the subsequent analysis for the estimation of R2∗ maps, while LEARN-BIO directly performs motion- and B0 -inhomogeneity-corrected R2∗ estimation. Both LEARN-IMG and LEARN-BIO jointly process all the available gradient echoes, which enables them to exploit spatial patterns available in the data. The high computational speed of LEARN-BIO is an advantage that can lead to a broader clinical application.

The Role of the Human Brain Neuron-Glia-Synapse Composition in Forming Resting-State Functional Connectivity Networks.

While significant progress has been achieved in studying resting-state functional networks in a healthy human brain and in a wide range of clinical conditions, many questions related to their relationship to the brain's cellular constituents remain. Here, we use quantitative Gradient-Recalled Echo (qGRE) MRI for mapping the human brain cellular composition and BOLD (blood-oxygen level-dependent) MRI to explore how the brain cellular constituents relate to resting-state functional networks. Results show that the BOLD signal-defined synchrony of connections between cellular circuits in network-defined individual functional units is mainly associated with the regional neuronal density, while the between-functional units' connectivity strength is also influenced by the glia and synaptic components of brain tissue cellular constituents. These mechanisms lead to a rather broad distribution of resting-state functional network properties. Visual networks with the highest neuronal density (but lowest density of glial cells and synapses) exhibit the strongest coherence of the BOLD signal as well as the strongest intra-network connectivity. The Default Mode Network (DMN) is positioned near the opposite part of the spectrum with relatively low coherence of the BOLD signal but with a remarkably balanced cellular contents, enabling DMN to have a prominent role in the overall organization of the brain and hierarchy of functional networks.

Deep learning using a biophysical model for robust and accelerated reconstruction of quantitative, artifact-free and denoised R2* images.

PURPOSE: To introduce a novel deep learning method for Robust and Accelerated Reconstruction (RoAR) of quantitative and B0-inhomogeneity-corrected R2* maps from multi-gradient recalled echo (mGRE) MRI data. METHODS: RoAR trains a convolutional neural network (CNN) to generate quantitative R2∗ maps free from field inhomogeneity artifacts by adopting a self-supervised learning strategy given (a) mGRE magnitude images, (b) the biophysical model describing mGRE signal decay, and (c) preliminary-evaluated F-function accounting for contribution of macroscopic B0 field inhomogeneities. Importantly, no ground-truth R2* images are required and F-function is only needed during RoAR training but not application. RESULTS: We show that RoAR preserves all features of R2* maps while offering significant improvements over existing methods in computation speed (seconds vs. hours) and reduced sensitivity to noise. Even for data with SNR = 5 RoAR produced R2* maps with accuracy of 22% while voxel-wise analysis accuracy was 47%. For SNR = 10 the RoAR accuracy increased to 17% vs. 24% for direct voxel-wise analysis. CONCLUSIONS: RoAR is trained to recognize the macroscopic magnetic field inhomogeneities directly from the input magnitude-only mGRE data and eliminate their effect on R2∗ measurements. RoAR training is based on the biophysical model and does not require ground-truth R2* maps. Since RoAR utilizes signal information not just from individual voxels but also accounts for spatial patterns of the signals in the images, it reduces the sensitivity of R2* maps to the noise in the data. These features plus high computational speed provide significant benefits for the potential usage of RoAR in clinical settings.