Medical Image Analysis Applications Research Articles

Integrating spatial configuration into heatmap regression based CNNs for landmark localization.

In many medical image analysis applications, only a limited amount of training data is available due to the costs of image acquisition and the large manual annotation effort required from experts. Training recent state-of-the-art machine learning methods like convolutional neural networks (CNNs) from small datasets is a challenging task. In this work on anatomical landmark localization, we propose a CNN architecture that learns to split the localization task into two simpler sub-problems, reducing the overall need for large training datasets. Our fully convolutional SpatialConfiguration-Net (SCN) learns this simplification due to multiplying the heatmap predictions of its two components and by training the network in an end-to-end manner. Thus, the SCN dedicates one component to locally accurate but ambiguous candidate predictions, while the other component improves robustness to ambiguities by incorporating the spatial configuration of landmarks. In our extensive experimental evaluation, we show that the proposed SCN outperforms related methods in terms of landmark localization error on a variety of size-limited 2D and 3D landmark localization datasets, i.e., hand radiographs, lateral cephalograms, hand MRIs, and spine CTs.

Evaluating reinforcement learning agents for anatomical landmark detection.

Automatic detection of anatomical landmarks is an important step for a wide range of applications in medical image analysis. Manual annotation of landmarks is a tedious task and prone to observer errors. In this paper, we evaluate novel deep reinforcement learning (RL) strategies to train agents that can precisely and robustly localize target landmarks in medical scans. An artificial RL agent learns to identify the optimal path to the landmark by interacting with an environment, in our case 3D images. Furthermore, we investigate the use of fixed- and multi-scale search strategies with novel hierarchical action steps in a coarse-to-fine manner. Several deep Q-network (DQN) architectures are evaluated for detecting multiple landmarks using three different medical imaging datasets: fetal head ultrasound (US), adult brain and cardiac magnetic resonance imaging (MRI). The performance of our agents surpasses state-of-the-art supervised and RL methods. Our experiments also show that multi-scale search strategies perform significantly better than fixed-scale agents in images with large field of view and noisy background such as in cardiac MRI. Moreover, the novel hierarchical steps can significantly speed up the searching process by a factor of 4-5 times.

Body region localization in whole-body low-dose CT images of PET/CT scans using virtual landmarks.

Radiological imaging and image interpretation for clinical decision making are mostly specific to each body region such as head and neck, thorax, abdomen, pelvis, and extremities. In this study, we present a new solution to trim automatically the given axial image stack into image volumes satisfying the given body region definition. The proposed approach consists of the following steps. First, a set of reference objects is selected and roughly segmented. Virtual landmarks (VLs) for the objects are then identified by using principal component analysis and recursive subdivision of the object via the principal axes system. The VLs can be defined based on just the binary objects or objects with gray values also considered. The VLs may lie anywhere with respect to the object, inside or outside, and rarely on the object surface, and are tethered to the object. Second, a classic neural network regressor is configured to learn the geometric mapping relationship between the VLs and the boundary locations of each body region. The trained network is then used to predict the locations of the body region boundaries. In this study, we focus on three body regions - thorax, abdomen, and pelvis, and predict their superior and inferior axial locations denoted by TS(I), TI(I), AS(I), AI(I), PS(I), and PI(I), respectively, for any given volume image I. Two kinds of reference objects - the skeleton and the lungs and airways, are employed to test the localization performance of the proposed approach. Our method is tested by using low-dose unenhanced computed tomography (CT) images of 180 near whole-body 18 F-fluorodeoxyglucose-positron emission tomography/computed tomography (FDG-PET/CT) scans (including 34 whole-body scans) which are randomly divided into training and testing sets with a ratio of 85%:15%. The procedure is repeated six times and three times for the case of lungs and skeleton, respectively, with different divisions of the entire data set at this proportion. For the case of using skeleton as a reference object, the overall mean localization error for the six locations expressed as number of slices (nS) and distance (dS) in mm, is found to be nS: 3.4, 4.7, 4.1, 5.2, 5.2, and 3.9; dS: 13.4, 18.9, 16.5, 20.8, 20.8, and 15.5mm for binary objects; nS: 4.1, 5.7, 4.3, 5.9, 5.9, and 4.0; dS: 16.2, 22.7, 17.2, 23.7, 23.7, and 16.1mm for gray objects, respectively. For the case of using lungs and airways as a reference object, the corresponding results are, nS: 4.0, 5.3, 4.1, 6.9, 6.9, and 7.4; dS: 15.0, 19.7, 15.3, 26.2, 26.2, and 27.9mm for binary objects; nS: 3.9, 5.4, 3.6, 7.2, 7.2, and 7.6; dS: 14.6, 20.1, 13.7, 27.3, 27.3, and 28.6mm for gray objects, respectively. Precise body region identification automatically in whole-body or body region tomographic images is vital for numerous medical image analysis and analytics applications. Despite its importance, this issue has received very little attention in the literature. We present a solution to this problem in this study using the concept of virtual landmarks. The method achieves localization accuracy within 2-3 slices, which is roughly comparable to the variation found in localization by experts. As long as the reference objects can be roughly segmented, the method with its learned VLs-to-boundary location relationship and predictive ability is transferable from one image modality to another.

PDE-constrained optimization in medical image analysis

PDE-constrained optimization problems find many applications in medical image analysis, for example, neuroimaging, cardiovascular imaging, and oncological imaging. We review related literature and give examples on the formulation, discretization, and numerical solution of PDE-constrained optimization problems for medical imaging. We discuss three examples. The first one is image registration. The second one is data assimilation for brain tumor patients, and the third one data assimilation in cardiovascular imaging. The image registration problem is a classical task in medical image analysis and seeks to find pointwise correspondences between two or more images. The data assimilation problems use a PDE-constrained formulation to link a biophysical model to patient-specific data obtained from medical images. The associated optimality systems turn out to be sets of nonlinear, multicomponent PDEs that are challenging to solve in an efficient way. The ultimate goal of our work is the design of inversion methods that integrate complementary data, and rigorously follow mathematical and physical principles, in an attempt to support clinical decision making. This requires reliable, high-fidelity algorithms with a short time-to-solution. This task is complicated by model and data uncertainties, and by the fact that PDE-constrained optimization problems are ill-posed in nature, and in general yield high-dimensional, severely ill-conditioned systems after discretization. These features make regularization, effective preconditioners, and iterative solvers that, in many cases, have to be implemented on distributed-memory architectures to be practical, a prerequisite. We showcase state-of-the-art techniques in scientific computing to tackle these challenges.

NiftyNet: a deep-learning platform for medical imaging

Background and objectivesMedical image analysis and computer-assisted intervention problems are increasingly being addressed with deep-learning-based solutions. Established deep-learning platforms are flexible but do not provide specific functionality for medical image analysis and adapting them for this domain of application requires substantial implementation effort. Consequently, there has been substantial duplication of effort and incompatible infrastructure developed across many research groups. This work presents the open-source NiftyNet platform for deep learning in medical imaging. The ambition of NiftyNet is to accelerate and simplify the development of these solutions, and to provide a common mechanism for disseminating research outputs for the community to use, adapt and build upon. MethodsThe NiftyNet infrastructure provides a modular deep-learning pipeline for a range of medical imaging applications including segmentation, regression, image generation and representation learning applications. Components of the NiftyNet pipeline including data loading, data augmentation, network architectures, loss functions and evaluation metrics are tailored to, and take advantage of, the idiosyncracies of medical image analysis and computer-assisted intervention. NiftyNet is built on the TensorFlow framework and supports features such as TensorBoard visualization of 2D and 3D images and computational graphs by default. ResultsWe present three illustrative medical image analysis applications built using NiftyNet infrastructure: (1) segmentation of multiple abdominal organs from computed tomography; (2) image regression to predict computed tomography attenuation maps from brain magnetic resonance images; and (3) generation of simulated ultrasound images for specified anatomical poses. ConclusionsThe NiftyNet infrastructure enables researchers to rapidly develop and distribute deep learning solutions for segmentation, regression, image generation and representation learning applications, or extend the platform to new applications.

Deep Learning Applications in Medical Image Analysis

The tremendous success of machine learning algorithms at image recognition tasks in recent years intersects with a time of dramatically increased use of electronic medical records and diagnostic imaging. This review introduces the machine learning algorithms as applied to medical image analysis, focusing on convolutional neural networks, and emphasizing clinical aspects of the field. The advantage of machine learning in an era of medical big data is that significant hierarchal relationships within the data can be discovered algorithmically without laborious hand-crafting of features. We cover key research areas and applications of medical image classification, localization, detection, segmentation, and registration. We conclude by discussing research obstacles, emerging trends, and possible future directions.

A Software Platform of Treatment Planning System for Radiotherapy Snake Robot

A software platform integrated of radiation treatment planning system and navigation system has been developed for a newly-developed snake robot. This platform is developed in a module-plugin form based on 3D Slicer medical image visualization and analysis application platform and an open-source toolkit SlicerRT. The platform realized function modules of image access and visualization, image segmentation, radiation path planning, treatment planning evaluation and optical navigation, and has provided the snake robot an extensible software experiment platform. It is beneficial for the brachytherapy research work and experiments.

End-to-End Adversarial Retinal Image Synthesis.

In medical image analysis applications, the availability of the large amounts of annotated data is becoming increasingly critical. However, annotated medical data is often scarce and costly to obtain. In this paper, we address the problem of synthesizing retinal color images by applying recent techniques based on adversarial learning. In this setting, a generative model is trained to maximize a loss function provided by a second model attempting to classify its output into real or synthetic. In particular, we propose to implement an adversarial autoencoder for the task of retinal vessel network synthesis. We use the generated vessel trees as an intermediate stage for the generation of color retinal images, which is accomplished with a generative adversarial network. Both models require the optimization of almost everywhere differentiable loss functions, which allows us to train them jointly. The resulting model offers an end-to-end retinal image synthesis system capable of generating as many retinal images as the user requires, with their corresponding vessel networks, by sampling from a simple probability distribution that we impose to the associated latent space. We show that the learned latent space contains a well-defined semantic structure, implying that we can perform calculations in the space of retinal images, e.g., smoothly interpolating new data points between two retinal images. Visual and quantitative results demonstrate that the synthesized images are substantially different from those in the training set, while being also anatomically consistent and displaying a reasonable visual quality.

Multi-Target Regression via Robust Low-Rank Learning.

Multi-target regression has recently regained great popularity due to its capability of simultaneously learning multiple relevant regression tasks and its wide applications in data mining, computer vision and medical image analysis, while great challenges arise from jointly handling inter-target correlations and input-output relationships. In this paper, we propose Multi-layer Multi-target Regression (MMR) which enables simultaneously modeling intrinsic inter-target correlations and nonlinear input-output relationships in a general framework via robust low-rank learning. Specifically, the MMR can explicitly encode inter-target correlations in a structure matrix by matrix elastic nets (MEN); the MMR can work in conjunction with the kernel trick to effectively disentangle highly complex nonlinear input-output relationships; the MMR can be efficiently solved by a new alternating optimization algorithm with guaranteed convergence. The MMR leverages the strength of kernel methods for nonlinear feature learning and the structural advantage of multi-layer learning architectures for inter-target correlation modeling. More importantly, it offers a new multi-layer learning paradigm for multi-target regression which is endowed with high generality, flexibility and expressive ability. Extensive experimental evaluation on 18 diverse real-world datasets demonstrates that our MMR can achieve consistently high performance and outperforms representative state-of-the-art algorithms, which shows its great effectiveness and generality for multivariate prediction.

Conditions on optimal support recovery in unmixing problems by means of multi-penalty regularization

Inspired by several real-life applications in audio processing and medical image analysis, where the quantity of interest is generated by several sources to be accurately modeled and separated, as well as by recent advances in regularization theory and optimization, we study the conditions on optimal support recovery in inverse problems of unmixing type by means of multi-penalty regularization. We consider and analyze a regularization functional composed of a data-fidelity term, where signal and noise are additively mixed, a non-smooth, convex, sparsity promoting term, and a quadratic penalty term to model the noise. We prove not only that the well-established theory for sparse recovery in the single parameter case can be translated to the multi-penalty settings, but we also demonstrate the enhanced properties of multi-penalty regularization in terms of support identification compared to sole -minimization. We additionally confirm and support the theoretical results by extensive numerical simulations, which give a statistics of robustness of the multi-penalty regularization scheme with respect to the single-parameter counterpart. Eventually, we confirm a significant improvement in performance compared to standard -regularization for compressive sensing problems considered in our experiments.

Combining deep learning and level set for the automated segmentation of the left ventricle of the heart from cardiac cine magnetic resonance.

We introduce a new methodology that combines deep learning and level set for the automated segmentation of the left ventricle of the heart from cardiac cine magnetic resonance (MR) data. This combination is relevant for segmentation problems, where the visual object of interest presents large shape and appearance variations, but the annotated training set is small, which is the case for various medical image analysis applications, including the one considered in this paper. In particular, level set methods are based on shape and appearance terms that use small training sets, but present limitations for modelling the visual object variations. Deep learning methods can model such variations using relatively small amounts of annotated training, but they often need to be regularised to produce good generalisation. Therefore, the combination of these methods brings together the advantages of both approaches, producing a methodology that needs small training sets and produces accurate segmentation results. We test our methodology on the MICCAI 2009 left ventricle segmentation challenge database (containing 15 sequences for training, 15 for validation and 15 for testing), where our approach achieves the most accurate results in the semi-automated problem and state-of-the-art results for the fully automated challenge.

Entropy-based particle correspondence for shape populations.

Statistical shape analysis of anatomical structures plays an important role in many medical image analysis applications such as understanding the structural changes in anatomy in various stages of growth or disease. Establishing accurate correspondence across object populations is essential for such statistical shape analysis studies. In this paper, we present an entropy-based correspondence framework for computing point-based correspondence among populations of surfaces in a groupwise manner. This robust framework is parameterization-free and computationally efficient. We review the core principles of this method as well as various extensions to deal effectively with surfaces of complex geometry and application-driven correspondence metrics. We apply our method to synthetic and biological datasets to illustrate the concepts proposed and compare the performance of our framework to existing techniques. Through the numerous extensions and variations presented here, we create a very flexible framework that can effectively handle objects of various topologies, multi-object complexes, open surfaces, and objects of complex geometry such as high-curvature regions or extremely thin features.

Low-Dimensional Non-Rigid Image Registration Using Statistical Deformation Models From Semi-Supervised Training Data.

Accurate and robust image registration is a fundamental task in medical image analysis applications, and requires non-rigid transformations with a large number of degrees of freedom. Statistical deformation models (SDMs) attempt to learn the distribution of non-rigid deformations, and can be used both to reduce the transformation dimensionality and to constrain the registration process. However, high-dimensional SDMs are difficult to train given orders of magnitude fewer training samples. In this paper, we utilize both a small set of annotated imaging data and a large set of unlabeled data to effectively learn an SDM of non-rigid transformations in a semi-supervised training (SST) framework. We demonstrate results applying this framework towards inter-subject registration of skull-stripped, magnetic resonance (MR) brain images. Our approach makes use of 39 labeled MR datasets to create a set of supervised registrations, which we augment with a set of over 1200 unsupervised registrations using unlabeled MRIs. Through leave-one-out cross validation, we show that SST of a non-rigid SDM results in a robust registration algorithm with significantly improved accuracy compared to standard, intensity-based registration, and does so with a 99% reduction in transformation dimensionality.

Computer-aided diagnosis from weak supervision: A benchmarking study

Supervised machine learning is a powerful tool frequently used in computer-aided diagnosis (CAD) applications. The bottleneck of this technique is its demand for fine grained expert annotations, which are tedious for medical image analysis applications. Furthermore, information is typically localized in diagnostic images, which makes representation of an entire image by a single feature set problematic. The multiple instance learning framework serves as a remedy to these two problems by allowing labels to be provided for groups of observations, called bags, and assuming the group label to be the maximum of the instance labels within the bag. This setup can effectively be applied to CAD by splitting a given diagnostic image into a Cartesian grid, treating each grid element (patch) as an instance by representing it with a feature set, and grouping instances belonging to the same image into a bag. We quantify the power of existing multiple instance learning methods by evaluating their performance on two distinct CAD applications: (i) Barrett's cancer diagnosis and (ii) diabetic retinopathy screening. In the experiments, mi-Graph appears as the best-performing method in bag-level prediction (i.e. diagnosis) for both of these applications that have drastically different visual characteristics. For instance-level prediction (i.e. disease localization), mi-SVM ranks as the most accurate method.

Hierarchical multi-atlas label fusion with multi-scale feature representation and label-specific patch partition

Multi-atlas patch-based label fusion methods have been successfully used to improve segmentation accuracy in many important medical image analysis applications. In general, to achieve label fusion a single target image is first registered to several atlas images. After registration a label is assigned to each target point in the target image by determining the similarity between the underlying target image patch (centered at the target point) and the aligned image patch in each atlas image. To achieve the highest level of accuracy during the label fusion process it's critical for the chosen patch similarity measurement to accurately capture the tissue/shape appearance of the anatomical structure. One major limitation of existing state-of-the-art label fusion methods is that they often apply a fixed size image patch throughout the entire label fusion procedure. Doing so may severely affect the fidelity of the patch similarity measurement, which in turn may not adequately capture complex tissue appearance patterns expressed by the anatomical structure. To address this limitation, we advance state-of-the-art by adding three new label fusion contributions: First, each image patch is now characterized by a multi-scale feature representation that encodes both local and semi-local image information. Doing so will increase the accuracy of the patch-based similarity measurement. Second, to limit the possibility of the patch-based similarity measurement being wrongly guided by the presence of multiple anatomical structures in the same image patch, each atlas image patch is further partitioned into a set of label-specific partial image patches according to the existing labels. Since image information has now been semantically divided into different patterns, these new label-specific atlas patches make the label fusion process more specific and flexible. Lastly, in order to correct target points that are mislabeled during label fusion, a hierarchical approach is used to improve the label fusion results. In particular, a coarse-to-fine iterative label fusion approach is used that gradually reduces the patch size. To evaluate the accuracy of our label fusion approach, the proposed method was used to segment the hippocampus in the ADNI dataset and 7.0T MR images, sub-cortical regions in LONI LBPA40 dataset, mid-brain regions in SATA dataset from MICCAI 2013 segmentation challenge, and a set of key internal gray matter structures in IXI dataset. In all experiments, the segmentation results of the proposed hierarchical label fusion method with multi-scale feature representations and label-specific atlas patches are more accurate than several well-known state-of-the-art label fusion methods.

Folding-Free Global Conformal Mapping for Genus-0 Surfaces by Harmonic Energy Minimization

Surface conformal maps between genus-0 surfaces play important roles in applied mathematics and engineering, with applications in medical image analysis and computer graphics. Previous work (Gu and Yau in Commun Inf Syst 2(2):121---146, 2002) introduces a variational approach, where global conformal parameterization of genus-0 surfaces was addressed through minimizing the harmonic energy, with two weaknesses: its gradient descent iteration is slow, and its solutions contain undesired parameterization foldings when the underlying surface has long sharp features. In this paper, we propose an algorithm that significantly accelerates the harmonic energy minimization and a method that iteratively removes foldings by taking advantages of the weighted Laplace---Beltrami eigen-projection. Experimental results show that the proposed approaches compute genus-0 surface harmonic maps much faster than the existing algorithm in Gu and Yau (Commun Inf Syst 2(2):121---146, 2002) and the new results contain no foldings.

Visual detection of knives in security applications using Active Appearance Models

In this paper, a novel application of Active Appearance Models to detecting knives in images is presented. In contrast to its popular applications in face segmentation and medical image analysis, we not only use this computer vision algorithm to locate an object that is known to exist in an analysed image, but---using an interest point typical of knives---also try to identify whether or not a knife exists in the image in question. We propose an entire detection scheme and examine its performance on a sample test set. The work presented in this paper aims to create a robust visual knife-detector to be used in security applications.

WESD--Weighted Spectral Distance for Measuring Shape Dissimilarity

This paper presents a new distance for measuring shape dissimilarity between objects. Recent publications introduced the use of eigenvalues of the Laplace operator as compact shape descriptors. Here, we revisit the eigenvalues to define a proper distance, called Weighted Spectral Distance (WESD), for quantifying shape dissimilarity. The definition of WESD is derived through analyzing the heat trace. This analysis provides the proposed distance with an intuitive meaning and mathematically links it to the intrinsic geometry of objects. We analyze the resulting distance definition, present and prove its important theoretical properties. Some of these properties include: 1) WESD is defined over the entire sequence of eigenvalues yet it is guaranteed to converge, 2) it is a pseudometric, 3) it is accurately approximated with a finite number of eigenvalues, and 4) it can be mapped to the [0,1) interval. Last, experiments conducted on synthetic and real objects are presented. These experiments highlight the practical benefits of WESD for applications in vision and medical image analysis.

Nonrigid Registration of Lung CT Images Based on Tissue Features

Nonrigid image registration is a prerequisite for various medical image process and analysis applications. Much effort has been devoted to thoracic image registration due to breathing motion. Recently, scale-invariant feature transform (SIFT) has been used in medical image registration and obtained promising results. However, SIFT is apt to detect blob features. Blobs key points are generally detected in smooth areas which may contain few diagnostic points. In general, diagnostic points used in medical image are often vessel crossing points, vascular endpoints, and tissue boundary points, which provide abundant information about vessels and can reflect the motion of lungs accurately. These points generally have high gradients as opposed to blob key points and can be detected by Harris. In this work, we proposed a hybrid feature detection method which can detect tissue features of lungs effectively based on Harris and SIFT. In addition, a novel method which can remove mismatched landmarks is also proposed. A series of thoracic CT images are tested by using the proposed algorithm, and the quantitative and qualitative evaluations show that our method is statistically significantly better than conventional SIFT method especially in the case of large deformation of lungs during respiration.

Value of a probabilistic atlas in medical image segmentation regarding non-rigid registration of abdominal CT scans

A probabilistic atlas provides important information to help segmentation and registration applications in medical image analysis. We construct a probabilistic atlas by picking a target geometry and mapping other training scans onto that target and then summing the results into one probabilistic atlas. By choosing an atlas space close to the desired target, we construct an atlas that represents the population well. Image registration used to map one image geometry onto another is a primary task in atlas building. One of the main parameters of registration is the choice of degrees of freedom (DOFs) of the geometric transform. Herein, we measure the effect of the registration’s DOFs on the segmentation performance of the resulting probabilistic atlas. Twenty-three normal abdominal CT scans were used, and four organs (liver, spinal cord, left and right kidneys) were segmented for each scan. A well-known manifold learning method, ISOMAP, was used to find the best target space to build an atlas. In summary, segmentation performance was high for high DOF registrations regardless of the chosen target space, while segmentation performance was lowered for low DOF registrations if a target space was far from the best target space. At the 0.05 level of statistical significance, there were no significant differences at high DOF registrations while there were significant differences at low DOF registrations when choosing different targets.