Learning Resolution-Adaptive Representations for Cross-Resolution Person Re-Identification.

Cross-resolution person re-identification (CRReID) is a challenging and practical problem that involves matching low-resolution (LR) query identity images against high-resolution (HR) gallery images. Query images often suffer from resolution degradation due to the different capturing conditions from real-world cameras. State-of-the-art solutions for CRReID either learn a resolution-invariant representation or adopt a super-resolution (SR) module to recover the missing information from the LR query. In this paper, we propose an alternative SR-free paradigm to directly compare HR and LR images via a dynamic metric that is adaptive to the resolution of a query image. We realize this idea by learning resolution-adaptive representations for cross-resolution comparison. We propose two resolution-adaptive mechanisms to achieve this. The first mechanism encodes the resolution specifics into different subvectors in the penultimate layer of the deep neural network, creating a varying-length representation. To better extract resolution-dependent information, we further propose to learn resolution-adaptive masks for intermediate residual feature blocks. A novel progressive learning strategy is proposed to train those masks properly. These two mechanisms are combined to boost the performance of CRReID. Experimental results show that the proposed method outperforms existing approaches and achieves state-of-the-art performance on multiple CRReID benchmarks.

Fully Connected Networks on a Diet With the Mediterranean Matrix Multiplication.

This article proposes the Mediterranean matrix multiplication, a new, simple and practical randomized algorithm that samples angles between the rows and columns of two matrices with sizes m, n, and p to approximate matrix multiplication in O(k(mn+np+mp)) steps, where k is a constant only related to the precision desired. The number of instructions carried out is mainly bounded by bitwise operators, amenable to a simplified processing architecture and compressed matrix weights. Results show that the method is superior in size and number of operations to the standard approximation with signed matrices. Equally important, this article demonstrates a first application to machine learning inference by showing that weights of fully connected layers can be compressed between 30 × and 100 × with little to no loss in inference accuracy. The requirements for pure floating-point operations are also down as our algorithm relies mainly on simpler bitwise operators.

3D Interactive Segmentation With Semi-Implicit Representation and Active Learning.

Segmenting complex 3D geometry is a challenging task due to rich structural details and complex appearance variations of target object. Shape representation and foreground-background delineation are two of the core components of segmentation. Explicit shape models, such as mesh based representations, suffer from poor handling of topological changes. On the other hand, implicit shape models, such as level-set based representations, have limited capacity for interactive manipulation. Fully automatic segmentation for separating foreground objects from background generally utilizes non-interoperable machine learning methods, which heavily rely on the off-line training dataset and are limited to the discrimination power of the chosen model. To address these issues, we propose a novel semi-implicit representation method, namely Non-Uniform Implicit B-spline Surface (NU-IBS), which adaptively distributes parametrically blended patches according to geometrical complexity. Then, a two-stage cascade classifier is introduced to carry out efficient foreground and background delineation, where a simplistic Naïve-Bayesian model is trained for fast background elimination, followed by a stronger pseudo-3D Convolutional Neural Network (CNN) multi-scale classifier to precisely identify the foreground objects. A localized interactive and adaptive segmentation scheme is incorporated to boost the delineation accuracy by utilizing the information iteratively gained from user intervention. The segmentation result is obtained via deforming an NU-IBS according to the probabilistic interpretation of delineated regions, which also imposes a homogeneity constrain for individual segments. The proposed method is evaluated on a 3D cardiovascular Computed Tomography Angiography (CTA) image dataset and Brain Tumor Image Segmentation Benchmark 2015 (BraTS2015) 3D Magnetic Resonance Imaging (MRI) dataset.

TLGP: a flexible transfer learning algorithm for gene prioritization based on heterogeneous source domain.

BACKGROUND: Gene prioritization (gene ranking) aims to obtain the centrality of genes, which is critical for cancer diagnosis and therapy since keys genes correspond to the biomarkers or targets of drugs. Great efforts have been devoted to the gene ranking problem by exploring the similarity between candidate and known disease-causing genes. However, when the number of disease-causing genes is limited, they are not applicable largely due to the low accuracy. Actually, the number of disease-causing genes for cancers, particularly for these rare cancers, are really limited. Therefore, there is a critical needed to design effective and efficient algorithms for gene ranking with limited prior disease-causing genes. RESULTS: In this study, we propose a transfer learning based algorithm for gene prioritization (called TLGP) in the cancer (target domain) without disease-causing genes by transferring knowledge from other cancers (source domain). The underlying assumption is that knowledge shared by similar cancers improves the accuracy of gene prioritization. Specifically, TLGP first quantifies the similarity between the target and source domain by calculating the affinity matrix for genes. Then, TLGP automatically learns a fusion network for the target cancer by fusing affinity matrix, pathogenic genes and genomic data of source cancers. Finally, genes in the target cancer are prioritized. The experimental results indicate that the learnt fusion network is more reliable than gene co-expression network, implying that transferring knowledge from other cancers improves the accuracy of network construction. Moreover, TLGP outperforms state-of-the-art approaches in terms of accuracy, improving at least 5%. CONCLUSION: The proposed model and method provide an effective and efficient strategy for gene ranking by integrating genomic data from various cancers.

Modeling Large Sparse Data for Feature Selection: Hospital Admission Predictions of the Dementia Patients Using Primary Care Electronic Health Records.

A growing elderly population suffering from incurable, chronic conditions such as dementia present a continual strain on medical services due to mental impairment paired with high comorbidity resulting in increased hospitalization risk. The identification of at risk individuals allows for preventative measures to alleviate said strain. Electronic health records provide opportunity for big data analysis to address such applications. Such data however, provides a challenging problem space for traditional statistics and machine learning due to high dimensionality and sparse data elements. This article proposes a novel machine learning methodology: entropy regularization with ensemble deep neural networks (ECNN), which simultaneously provides high predictive performance of hospitalization of patients with dementia whilst enabling an interpretable heuristic analysis of the model architecture, able to identify individual features of importance within a large feature domain space. Experimental results on health records containing 54,647 features were able to identify 10 event indicators within a patient timeline: a collection of diagnostic events, medication prescriptions and procedural events, the highest ranked being essential hypertension. The resulting subset was still able to provide a highly competitive hospitalization prediction (Accuracy: 0.759) as compared to the full feature domain (Accuracy: 0.755) or traditional feature selection techniques (Accuracy: 0.737), a significant reduction in feature size. The discovery and heuristic evidence of correlation provide evidence for further clinical study of said medical events as potential novel indicators. There also remains great potential for adaption of ECNN within other medical big data domains as a data mining tool for novel risk factor identification.

Harnessing the Power of Machine Learning in Dementia Informatics Research: Issues, Opportunities, and Challenges.

Dementia is a chronic and degenerative condition affecting millions globally. The care of patients with dementia presents an ever-continuing challenge to healthcare systems in the 21st century. Medical and health sciences have generated unprecedented volumes of data related to health and wellbeing for patients with dementia due to advances in information technology, such as genetics, neuroimaging, cognitive assessment, free texts, routine electronic health records, etc. Making the best use of these diverse and strategic resources will lead to high-quality care of patients with dementia. As such, machine learning becomes a crucial factor in achieving this objective. The aim of this paper is to provide a state-of-the-art review of machine learning methods applied to health informatics for dementia care. We collate and review the existing scientific methodologies and identify the relevant issues and challenges when faced with big health data. Machine learning has demonstrated promising applications to neuroimaging data analysis for dementia care, while relatively less effort has been made to make use of integrated heterogeneous data via advanced machine learning approaches. We further indicate future potential and research directions in applying advanced machine learning, such as deep learning, to dementia informatics.

Coupled s-excess HMM for vessel border tracking and segmentation.

In this paper, we present a novel image segmentation technique, based on hidden Markov model (HMM), which we then apply to simultaneously segment interior and exterior walls of fluorescent confocal images of lymphatic vessels. Our proposed method achieves this by tracking hidden states, which are used to indicate the locations of both the inner and outer wall borders throughout the sequence of images. We parameterize these vessel borders using radial basis functions (RBFs), thus enabling us to minimize the number of points we need to track as we progress through multiple layers and therefore reduce computational complexity. Information about each border is detected using patch-wise convolutional neural networks (CNN). We use the softmax function to infer the emission probability and use a proposed new training algorithm based on s-excess optimization to learn the transition probability. We also introduce a new optimization method to determine the optimum sequence of the hidden states. Thus, we transform the segmentation problem into one that minimizes an s-excess graph cut, where each hidden state is represented as a graph node and the weight of these nodes are defined by their emission probabilities. The transition probabilities are used to define relationships between neighboring nodes in the constructed graph. We compare our proposed method to the Viterbi and Baum-Welch algorithms. Both qualitative and quantitative analysis show superior performance of the proposed methods.

Phase contrast cell detection using multilevel classification.

In this paper, we propose a fully automated learning-based approach for detecting cells in time-lapse phase contrast images. The proposed system combines 2 machine learning approaches to achieve bottom-up image segmentation. We apply pixel-wise classification using random forests (RF) classifiers to determine the potential location of the cells. Each pixel is classified into 4 categories (cell, mitotic cell, halo effect, and background noise). Various image features are extracted at different scales to train the RF classifier. The resulting probability map is partitioned using the k-means algorithm to form potential cell regions. These regions are expanded into the neighboring areas to recover some missing or broken cell regions. To validate the cell regions, another machine learning method based on the bag-of-features and spatial pyramid encoding is proposed. The result of the second classifier can be a validated cell, a merged cell, or a noncell. In the case that the cell region is classified as a merged cell, it is split by using the seeded watershed method. The proposed method is demonstrated on several phase contrast image datasets, ie, U2OS, HeLa, and NIH 3T3. In comparison to state-of-the-art cell detection techniques, the proposed method shows improved performance, particularly in dealing with noise interference and drastic shape variations.

Estimating the accuracy of a reduced-order model for the calculation of fractional flow reserve (FFR).

Image-based noninvasive fractional flow reserve (FFR) is an emergent approach to determine the functional relevance of coronary stenoses. The present work aimed to determine the feasibility of using a method based on coronary computed tomography angiography (CCTA) and reduced-order models (0D-1D) for the evaluation of coronary stenoses. The reduced-order methodology (cFFRRO ) was kept as simple as possible and did not include pressure drop or stenosis models. The geometry definition was incorporated into the physical model used to solve coronary flow and pressure. cFFRRO was assessed on a virtual cohort of 30 coronary artery stenoses in 25 vessels and compared with a standard approach based on 3D computational fluid dynamics (cFFR3D ). In this proof-of-concept study, we sought to investigate the influence of geometry and boundary conditions on the agreement between both methods. Performance on a per-vessel level showed a good correlation between both methods (Pearson's product-moment R=0.885, P<0.01), when using cFFR3D  as the reference standard. The 95% limits of agreement were -0.116 and 0.08, and the mean bias was -0.018 (SD =0.05). Our results suggest no appreciable difference between cFFRRO  and cFFR3D with respect to lesion length and/or aspect ratio. At a fixed aspect ratio, however, stenosis severity and shape appeared to be the most critical factors accounting for differences in both methods. Despite the assumptions inherent to the 1D formulation, asymmetry did not seem to affect the agreement. The choice of boundary conditions is critical in obtaining a functionally significant drop in pressure. Our initial data suggest that this approach may be part of a broader risk assessment strategy aimed at increasing the diagnostic yield of cardiac catheterisation for in-hospital evaluation of haemodynamically significant stenoses.

Energy minimization in medical image analysis: Methodologies and applications.

Energy minimization is of particular interest in medical image analysis. In the past two decades, a variety of optimization schemes have been developed. In this paper, we present a comprehensive survey of the state-of-the-art optimization approaches. These algorithms are mainly classified into two categories: continuous method and discrete method. The former includes Newton-Raphson method, gradient descent method, conjugate gradient method, proximal gradient method, coordinate descent method, and genetic algorithm-based method, while the latter covers graph cuts method, belief propagation method, tree-reweighted message passing method, linear programming method, maximum margin learning method, simulated annealing method, and iterated conditional modes method. We also discuss the minimal surface method, primal-dual method, and the multi-objective optimization method. In addition, we review several comparative studies that evaluate the performance of different minimization techniques in terms of accuracy, efficiency, or complexity. These optimization techniques are widely used in many medical applications, for example, image segmentation, registration, reconstruction, motion tracking, and compressed sensing. We thus give an overview on those applications as well.

Divergence of Gradient Convolution: Deformable Segmentation With Arbitrary Initializations.

In this paper, we propose a unified approach to deformable model-based segmentation. The fundamental force field of the proposed method is based on computing the divergence of a gradient convolution field (GCF), which makes the full use of directional information of the image gradient vectors and their interactions across image domain. However, instead of directly using such a vector field for deformable segmentation as in the conventional approaches, we derive a more salient representation for contour evolution, and very importantly, we demonstrate that this representation of image force field not only leads to global minimum through convex relaxation but also can achieve the same result using the conventional gradient descent with an intrinsic regularization. Thus, the proposed method can handle arbitrary initializations. The proposed external force field for deformable segmentation has both edge-based properties in that the GCF is computed from image gradients, and the region-based attributes since its divergence can be treated as a region indication function. Moreover, nonlinear diffusion can be conveniently applied to GCF to improve its performance in dealing with noise interference. We also show the extension of GCF from 2D to 3D. In comparison to the state-of-the-art deformable segmentation techniques, the proposed method shows greater flexibility in model initialization and optimization realization, as well as better performance toward noise interference and appearance variation.

Automatic segmentation of lymph vessel wall using optimal surface graph cut and hidden Markov Models.

We present a novel method to segment the lymph vessel wall in confocal microscopy images using Optimal Surface Segmentation (OSS) and hidden Markov Models (HMM). OSS is used to preform a pre-segmentation on the images, to act as the initial state for the HMM. We utilize a steerable filter to determine edge based filters for both of these segmentations, and use these features to build Gaussian probability distributions for both the vessel walls and the background. From this we infer the emission probability for the HMM, and the transmission probability is learned using a Baum-Welch algorithm. We transform the segmentation problem into one of cost minimization, with each node in the graph corresponding to one state, and the weight for each node being defined using its emission probability. We define the inter-relations between neighboring nodes using the transmission probability. Having constructed the problem, it is solved using the Viterbi algorithm, allowing the vessel to be reconstructed. The optimal solution can be found in polynomial time. We present qualitative and quantitative analysis to show the performance of the proposed method.

A multi-stage random forest classifier for phase contrast cell segmentation.

We present a machine learning based approach to automatically detect and segment cells in phase contrast images. The proposed method consists of a multi-stage classification scheme based on random forest (RF) classifier. Both low level and mid level image features are used to determine meaningful cell regions. Pixel-wise RF classification is first carried out to categorize pixels into 4 classes (dark cell, bright cell, halo artifact, and background) and generate a probability map for cell regions. K-means clustering is then applied on the probability map to group similar pixels into candidate cell regions. Finally, cell validation is performed by another RF to verify the candidate cell regions. The proposed method has been tested on U2-OS human osteosarcoma phase contrast images. The experimental results show better performance of the proposed method with precision 92.96% and recall 96.63% compared to a state-of-the-art segmentation technique.

Computer Vision Techniques for Transcatheter Intervention.

Minimally invasive transcatheter technologies have demonstrated substantial promise for the diagnosis and the treatment of cardiovascular diseases. For example, transcatheter aortic valve implantation is an alternative to aortic valve replacement for the treatment of severe aortic stenosis, and transcatheter atrial fibrillation ablation is widely used for the treatment and the cure of atrial fibrillation. In addition, catheter-based intravascular ultrasound and optical coherence tomography imaging of coronary arteries provides important information about the coronary lumen, wall, and plaque characteristics. Qualitative and quantitative analysis of these cross-sectional image data will be beneficial to the evaluation and the treatment of coronary artery diseases such as atherosclerosis. In all the phases (preoperative, intraoperative, and postoperative) during the transcatheter intervention procedure, computer vision techniques (e.g., image segmentation and motion tracking) have been largely applied in the field to accomplish tasks like annulus measurement, valve selection, catheter placement control, and vessel centerline extraction. This provides beneficial guidance for the clinicians in surgical planning, disease diagnosis, and treatment assessment. In this paper, we present a systematical review on these state-of-the-art methods. We aim to give a comprehensive overview for researchers in the area of computer vision on the subject of transcatheter intervention. Research in medical computing is multi-disciplinary due to its nature, and hence, it is important to understand the application domain, clinical background, and imaging modality, so that methods and quantitative measurements derived from analyzing the imaging data are appropriate and meaningful. We thus provide an overview on the background information of the transcatheter intervention procedures, as well as a review of the computer vision techniques and methodologies applied in this area.

Combining region-based and imprecise boundary-based cues for interactive medical image segmentation.

In this paper, we present an approach combining both region selection and user point selection for user-assisted segmentation as either an enclosed object or an open curve, investigate the method of image segmentation in specific medical applications (user-assisted segmentation of the media-adventitia border in intravascular ultrasound images, and lumen border in optical coherence tomography images), and then demonstrate the method with generic images to show how it could be utilized in other types of medical image and is not limited to the applications described. The proposed method combines point-based soft constraint on object boundary and stroke-based regional constraint. The user points act as attraction points and are treated as soft constraints rather than hard constraints that the segmented boundary has to pass through. The user can also use strokes to specify region of interest. The probabilities of region of interest for each pixel are then calculated, and their discontinuity is used to indicate object boundary. The combinations of different types of user constraints and image features allow flexible and robust segmentation, which is formulated as an energy minimization problem on a multilayered graph and is solved using a shortest path search algorithm. We show that this combinatorial approach allows efficient and effective interactive segmentation, which can be used with both open and closed curves to segment a variety of images in different ways. The proposed method is demonstrated in the two medical applications, that is, intravascular ultrasound and optical coherence tomography images, where image artefacts such as acoustic shadow and calcification are commonplace and thus user guidance is desirable. We carried out both qualitative and quantitative analysis of the results for the medical data; comparing the proposed method against a number of interactive segmentation techniques.

Segmentation of biomedical images using active contour model with robust image feature and shape prior.

In this article, a new level set model is proposed for the segmentation of biomedical images. The image energy of the proposed model is derived from a robust image gradient feature which gives the active contour a global representation of the geometric configuration, making it more robust in dealing with image noise, weak edges, and initial configurations. Statistical shape information is incorporated using nonparametric shape density distribution, which allows the shape model to handle relatively large shape variations. The segmentation of various shapes from both synthetic and real images depict the robustness and efficiency of the proposed method.

Integrated segmentation and interpolation of sparse data.

We address the two inherently related problems of segmentation and interpolation of 3D and 4D sparse data and propose a new method to integrate these stages in a level set framework. The interpolation process uses segmentation information rather than pixel intensities for increased robustness and accuracy. The method supports any spatial configurations of sets of 2D slices having arbitrary positions and orientations. We achieve this by introducing a new level set scheme based on the interpolation of the level set function by radial basis functions. The proposed method is validated quantitatively and/or subjectively on artificial data and MRI and CT scans and is compared against the traditional sequential approach, which interpolates the images first, using a state-of-the-art image interpolation method, and then segments the interpolated volume in 3D or 4D. In our experiments, the proposed framework yielded similar segmentation results to the sequential approach but provided a more robust and accurate interpolation. In particular, the interpolation was more satisfactory in cases of large gaps, due to the method taking into account the global shape of the object, and it recovered better topologies at the extremities of the shapes where the objects disappear from the image slices. As a result, the complete integrated framework provided more satisfactory shape reconstructions than the sequential approach.

Automatic bootstrapping and tracking of object contours.

A new fully automatic object tracking and segmentation framework is proposed. The framework consists of a motion-based bootstrapping algorithm concurrent to a shape-based active contour. The shape-based active contour uses finite shape memory that is automatically and continuously built from both the bootstrap process and the active-contour object tracker. A scheme is proposed to ensure that the finite shape memory is continuously updated but forgets unnecessary information. Two new ways of automatically extracting shape information from image data given a region of interest are also proposed. Results demonstrate that the bootstrapping stage provides important motion and shape information to the object tracker. This information is found to be essential for good (fully automatic) initialization of the active contour. Further results also demonstrate convergence properties of the content of the finite shape memory and similar object tracking performance in comparison with an object tracker with unlimited shape memory. Tests with an active contour using a fixed-shape prior also demonstrate superior performance for the proposed bootstrapped finite-shape-memory framework and similar performance when compared with a recently proposed active contour that uses an alternative online learning model.

Geometrically induced force interaction for three-dimensional deformable models.

In this paper, we propose a novel 3-D deformable model that is based upon a geometrically induced external force field which can be conveniently generalized to arbitrary dimensions. This external force field is based upon hypothesized interactions between the relative geometries of the deformable model and the object boundary characterized by image gradient. The evolution of the deformable model is solved using the level set method so that topological changes are handled automatically. The relative geometrical configurations between the deformable model and the object boundaries contribute to a dynamic vector force field that changes accordingly as the deformable model evolves. The geometrically induced dynamic interaction force has been shown to greatly improve the deformable model performance in acquiring complex geometries and highly concave boundaries, and it gives the deformable model a high invariancy in initialization configurations. The voxel interactions across the whole image domain provide a global view of the object boundary representation, giving the external force a long attraction range. The bidirectionality of the external force field allows the new deformable model to deal with arbitrary cross-boundary initializations, and facilitates the handling of weak edges and broken boundaries. In addition, we show that by enhancing the geometrical interaction field with a nonlocal edge-preserving algorithm, the new deformable model can effectively overcome image noise. We provide a comparative study on the segmentation of various geometries with different topologies from both synthetic and real images, and show that the proposed method achieves significant improvements against existing image gradient techniques.

Active contouring based on gradient vector interaction and constrained level set diffusion.

This paper presents an extension of our recently introduced MAC model to deal with the initialization dependency problem that commonly appears in edge-based approaches. Its dynamic force field, unique bidirectionality, and constrained diffusion-based level set evolution provide great freedom in contour initialization and show significant improvements in initialization independency compared to other edge-based techniques. It can handle more sophisticated topological changes than splitting and merging. It provides new potentials for edge-based active contour methods, particularly when detecting and localizing objects with unknown location, geometry, and topology.