Compressive Sensing for Magnetic Resonance Imaging (CS-MRI) aims to reconstruct Magnetic Resonance (MR) images from under-sampled raw data. There are two challenges to improve CS-MRI methods, i.e. designing an under-sampling algorithm to achieve optimal sampling, as well as designing fast and small deep neural networks to obtain reconstructed MR images with superior quality. To improve the reconstruction quality of MR images, we propose a novel deep convolutional neural network architecture for CS-MRI named MRCSNet. The MRCSNet consists of three sub-networks, a compressive sensing sampling sub-network, an initial reconstruction sub-network, and a refined reconstruction sub-network. Experimental results demonstrate that MRCSNet generates high-quality reconstructed MR images at various under-sampling ratios, and also meets the requirements of real-time CS-MRI applications. Compared to state-of-the-art CS-MRI approaches, MRCSNet offers a significant improvement in reconstruction accuracies, such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM). Besides, it reduces the reconstruction error evaluated by the Normalized Root-Mean-Square Error (NRMSE). The source codes are available at https://github.com/TaihuLight/MRCSNet .

As the majority of popular deep neural network (DNN) frameworks focus on a closed format CUDA implementations based on one or more NVIDIA GPUs, they cannot efficiently leverage other devices in cluster mode to accelerate the training and inference of DNNs except NVIDIA GPUs. To accelerate DNNs using heterogeneous multi-/many-core clusters, we propose an OpenCL-based DNN framework called UHCL-Darknet. First, we design a unified OpenCL platform model for the heterogeneous cluster called UHCL, and an adaptive runtime system with the affinity-based dynamic scheduler for UHCL, enabling transparent utilization of a wide variety of vendor-specific OpenCL devices in the heterogeneous cluster. Then, we extend Darknet to UHCL by introducing the parallel optimization of DNNs, such as paralleling Winogrand-based convolutions and auto-tuning parameterized OpenCL kernels. The training and inference of art-of-data DNN models (e.g., YOLOv2, ResNet-50, and DenseNet-201) are executed on an experimental heterogeneous cluster. Results show that UHCL-Darknet is a scalable and portable DNN framework for heterogeneous clusters, and achieves 1.9x and 2.2x speedups on average respectively for the image throughput of data-parallel training and inference on the experimental heterogeneous cluster.

Clustering is an effective method for data analysis and can be exploited to unknown features of data samples, its applications range from data mining to bioinformatics analysis. Several clustering approaches have been proposed in order to obtain a better trade-off between accuracy and efficiency of the clustering process. It is well-known that no existing clustering algorithm completely satisfies both accuracy and efficiency requirements, thus we propose a clustering algorithm called ADCMK (for Automatic Density Clustering with Multiple Kernels) exhibiting higher quality than the density ones proposed so far, while allowing users to cluster efficiently without determining parameters manually. The algorithm consists of learning optimal combined kernel, reducing dimensionality with the optimal kernel, automatically detecting cluster centroids with outliers test, assigning clusters and visualizing results. The proposed clustering algorithm is extensively tested on several well-known high-dimensional dataset in biomedical and bioinformatics field. The results show that the new algorithm tends to automatically produce clusters of better quality than other density clustering algorithms.