Acute ischemic stroke is recognized as a common cerebral vascular disease in aging people. Accurate diagnosis and timely treatment can effectively improve the blood supply of the ischemic area and reduce the risk of disability or even death. Understanding the location and size of infarcts plays a critical role in the diagnosis decision. However, manual localization and quantification of stroke lesions are laborious and time-consuming. In this paper, we propose a novel automatic method to segment acute ischemic stroke from diffusion weighted images (DWIs) using deep 3-D convolutional neural networks (CNNs). Our method can efficiently utilize 3-D contextual information and automatically learn very discriminative features in an end-to-end and data-driven way. To relieve the difficulty of training very deep 3-D CNN, we equip our network with dense connectivity to enable the unimpeded propagation of information and gradients throughout the network. We train our model with Dice objective function to combat the severe class imbalance problem in data. A DWI data set containing 242 subjects (90 for training, 62 for validation, and 90 for testing) with various types of acute ischemic stroke was constructed to evaluate our method. Our model achieved high performance on various metrics (Dice similarity coefficient: 79.13%, lesionwise precision: 92.67%, and lesionwise F1 score: 89.25%), outperforming the other state-of-the-art CNN methods by a large margin. We also evaluated the model on ISLES2015-SSIS data set and achieved very competitive performance, which further demonstrated its generalization capacity. The proposed method is fast and accurate, demonstrating a good potential in clinical routines.

Magnetic resonance imaging (MRI) has been extensively used in clinical practice but suffers from long data acquisition time. Following the success of compressed sensing (CS) theory, many efforts have been made to accurately reconstruct MR images from undersampled k-space measurements and therefore dramatically reduce MRI scan time. To further improve image quality, we formulate undersampled MRI reconstruction as a least-squares optimization problem regularized by shearlet transform and overlapping-group sparsity-promoting total variation (OSTV). Shearlet transform, a directional representation system, is capable of capturing the optimal sparse representation for images with plentiful geometrical information. OSTV performs well in suppressing staircase-like artifacts often arising in traditional TV-based reconstructed images. To guarantee solution stability and efficiency, the resulting optimization problem is solved using an alternating direction methods of multipliers (ADMM)-based numerical algorithm. Extensive experimental results on both phantom and in vivo MRI datasets have demonstrated the superior performance of our proposed method in terms of both quantitative evaluation and visual quality.

Cerebral microbleeds (CMBs) are small haemorrhages nearby blood vessels. They have been recognized as important diagnostic biomarkers for many cerebrovascular diseases and cognitive dysfunctions. In current clinical routine, CMBs are manually labelled by radiologists but this procedure is laborious, time-consuming, and error prone. In this paper, we propose a novel automatic method to detect CMBs from magnetic resonance (MR) images by exploiting the 3D convolutional neural network (CNN). Compared with previous methods that employed either low-level hand-crafted descriptors or 2D CNNs, our method can take full advantage of spatial contextual information in MR volumes to extract more representative high-level features for CMBs, and hence achieve a much better detection accuracy. To further improve the detection performance while reducing the computational cost, we propose a cascaded framework under 3D CNNs for the task of CMB detection. We first exploit a 3D fully convolutional network (FCN) strategy to retrieve the candidates with high probabilities of being CMBs, and then apply a well-trained 3D CNN discrimination model to distinguish CMBs from hard mimics. Compared with traditional sliding window strategy, the proposed 3D FCN strategy can remove massive redundant computations and dramatically speed up the detection process. We constructed a large dataset with 320 volumetric MR scans and performed extensive experiments to validate the proposed method, which achieved a high sensitivity of 93.16% with an average number of 2.74 false positives per subject, outperforming previous methods using low-level descriptors or 2D CNNs by a significant margin. The proposed method, in principle, can be adapted to other biomarker detection tasks from volumetric medical data.

Enhancement of vessels in medical images is still an unsolved problem. Multiscale approaches were proposed to improve the vessel enhancement effect based on the structure size and image resolution. Vessel enhancing diffusion (VED) filter is one of the multiscale approaches, which was based on the scale space theory. VED performs well on enhancing vessel structures but cannot preserve complex structures such as the vessel junctions. In this paper, a structure-preserving diffusion tensor is defined in the diffusion equation, which brings a structure-preserving vessel enhancing diffusion filter. Through the multiscale framework, the proposed method enhances the vessel structures especially the complex structure such as junctions. Experimental evaluation performed on various vessel data sets demonstrated the effectiveness of the proposed method.

High-resolution peripheral quantitative computed tomography (HR-pQCT) is a new noninvasive bone imaging technology that generates high-resolution 3-D images for quantitatively analysis of the bone microarchitecture in human. To enable quantitative evaluation of bone changes, either bone gain or loss, accurate alignment between the baseline and follow-up scans of the same individual is necessary. The major difficulties in achieving efficient and automatic registration of the HR-pQCT data are the large data size, deformations in the nonskeletal structures, and the complexity of the trabecular bone geometry. In this paper, we propose an automatic surface-based approach for fast and accurate registration of the HR-pQCT data, where the rigid registration is applied on the surfaces of the bony structures extracted from the grayscale HR-pQCT. The bony structure segmentation is performed via an automatic method that can adaptively determine the thresholds for separating the bony structure from the background and nonskeletal tissues. Experimental results performed on ten pairs of baseline and follow-up wrist scans of five adolescents and five elderly patients with osteoporosis showed the advantage of the proposed method in the high degree of automation, while the resultant parameters describing bone mineral density and trabecular architecture after registration were comparable with the outputs of the scanner's software. This automatic and accurate matching procedure may contribute to the clinical application and research of HR-pQCT.

The support vector machine (SVM) has been demonstrated to be a very effective classifier in many applications, but its performance is still limited as the data distribution information is underutilized in determining the decision hyperplane. Most of the existing kernels employed in nonlinear SVMs measure the similarity between a pair of pattern images based on the Euclidean inner product or the Euclidean distance of corresponding input patterns, which ignores data distribution tendency and makes the SVM essentially a "local" classifier. In this paper, we provide a step toward a paradigm of kernels by incorporating data specific knowledge into existing kernels. We first find the data structure for each class adaptively in the input space via agglomerative hierarchical clustering (AHC), and then construct the weighted Mahalanobis distance (WMD) kernels using the detected data distribution information. In WMD kernels, the similarity between two pattern images is determined not only by the Mahalanobis distance (MD) between their corresponding input patterns but also by the sizes of the clusters they reside in. Although WMD kernels are not guaranteed to be positive definite (pd) or conditionally positive definite (cpd), satisfactory classification results can still be achieved because regularizers in SVMs with WMD kernels are empirically positive in pseudo-Euclidean (pE) spaces. Experimental results on both synthetic and real-world data sets show the effectiveness of "plugging" data structure into existing kernels.

The generalization error bounds found by current error models using the number of effective parameters of a classifier and the number of training samples are usually very loose. These bounds are intended for the entire input space. However, support vector machine (SVM), radial basis function neural network (RBFNN), and multilayer perceptron neural network (MLPNN) are local learning machines for solving problems and treat unseen samples near the training samples to be more important. In this paper, we propose a localized generalization error model which bounds from above the generalization error within a neighborhood of the training samples using stochastic sensitivity measure. It is then used to develop an architecture selection technique for a classifier with maximal coverage of unseen samples by specifying a generalization error threshold. Experiments using 17 University of California at Irvine (UCI) data sets show that, in comparison with cross validation (CV), sequential learning, and two other ad hoc methods, our technique consistently yields the best testing classification accuracy with fewer hidden neurons and less training time.

Large margin classifiers have been widely applied in solving supervised learning problems. One representative model in large margin learning is the support vector machine (SVM). SVM is an unstructured classifier since the data structure information is underutilized and the decision hyperplane calculation relies exclusively on the support vectors. To incorporate the data covariance information into the large margin learning, structured large margin machine (SLMM) is recently proposed and show better performance than classical SVM in some applications. Instead of utilizing the data structures straightly like SLMM, SVM ensemble (SVMe) improves the generalization ability of SVM in another way by combining the outputs of a series of SVMs. Inspired by SVMe, we are going to explore the ensemble counterpart for SLMM, i.e., SLMMe, and validate the effectiveness of multiple SLMM system. Experimental results on benchmark datasets demonstrate that SLMMe improves SLMM by reducing its variance, and SLMMe outperforms SVMe in most cases in terms of both classification accuracy and variance.

The one-class classification problem aims to distinguish a target class from outliers. The spherical one-class classifier (SOCC) solves this problem by finding a hypersphere with minimum volume that contains the target data while keeping outlier samples outside. SOCC achieves satisfactory performance only when the target samples have the same distribution tendency in all orientations. Therefore, the performance of the SOCC is limited in the way that many superfluous outliers might be mistakenly enclosed. The authors propose to exploit target data structures obtained via unsupervised methods such as agglomerative hierarchical clustering and use them in calculating a set of hyperellipsoidal separating boundaries. This method is named the structured one-class classifier (TOCC). The optimization problem in TOCC can be formulated as a series of second-order cone programming problems that can be solved with acceptable efficiency by primal-dual interior-point methods. The experimental results on artificially generated data sets and benchmark data sets demonstrate the advantages of TOCC.