Learning to see stuff

vertebral fracture is a very common outcome of osteoporosis, which is one of the major public health concerns in the world. Early detection of vertebral fractures is important because timely pharmacologic intervention can reduce the risk of subsequent additional fractures. Our goal seeks to develop a computerized method for detection of vertebral fractures by measuring the shape and appearance of vertebrae on cervical xray radiographs in order to assist radiologist’s image interpretation and thus allow the early diagnosis of osteoporosis. The statistical models of shape and appearance are powerful tools for interpreting medical images. This work introduces the application of correlation filter classifiers for identification and verification of the osteoporosis presence in cervical vertebrae training/ testing set. Correlation filter classifiers have been previously applied to other biometric classification tasks, but not to classification of cervical vertebrae images. We describe how the extraction of an appropriate region of interest in the cervical vertebrae surface can be used to design correlation filters that accomplish 90 % recognition on a database of 50 cervical bone shapes.Keywordsx-ray radiographsSegmentationASM modelingCorrelation filtervertebral deformityOsteoporosis

The joint effect of mesoscale and microscale roughness on perceived gloss

Active Appearance Models (AAMs) and Constrained Local Models (CLMs) are classical approaches for model-based image segmentation in medical image analysis. AAMs consist of global statistical models of shape and appearance, are known to be hard to fit, and often suffer from insufficient generalization capabilities in case of limited training data. CLMs model appearance only for local patches and relax or completely remove the global appearance constraint. They are, therefore, much easier to optimize but in certain cases they lack the robustness of AAMs. In this paper, we present a framework for patch-based active appearance modeling, which elegantly combines strengths of AAMs and CLMs. Our models provide global shape and appearance constraints and we make use of recent methodological advances from computer vision for efficient joint optimization of shape and appearance parameters during model fitting. Furthermore, the insufficient generalization abilities of those global models are tackled by incorporating and extending a recent approach for learning representative statistical shape models from small training populations. We evaluate our approach on publicly available chest radiographs and cardiac MRI data. The results show that the proposed framework leads to competitive results in terms of segmentation accuracy for challenging multi-object segmentation problems even when only few training samples are available.

Fractures of the proximal femur are one of the principal causes of mortality among elderly persons. Traditional methods for the determination of femoral fracture risk use methods for measuring bone mineral density. However, BMD alone is not sufficient to predict bone failure load for an individual patient and additional parameters have to be determined for this purpose. In this work an approach that uses statistical models of appearance to identify relevant regions and parameters for the prediction of biomechanical properties of the proximal femur will be presented. By using Support Vector Regression the proposed model based approach is capable of predicting two different biomechanical parameters accurately and fully automatically in two different testing scenarios.

The Active Appearance Model (AAM) algorithm has proved to be a successful method for matching statistical models of appearance to new images. Since the original algorithm was described there have been a variety of suggested modifications to the basic algorithm, each typically claiming to be in some way superior. We review these algorithms and report the results of experiments comparing their performance. We also investigate the effects of different methods of estimating the update matrix used in the algorithm. We find that careful choice of the latter has at least as much effect as the choice of updating technique.

In this paper, we propose a method for joint 2D segmentation and 2D-3D pose tracking. First, we define a novel energy functional which considers the discrimination between statistical appearance models and the coherence among neighboring pixels simultaneously. And then, a particle filter-like stochastic optimization technique is adopted to solve the energy functional, so that a preferable initial value can be provided for the subsequent damped Newton optimization method. Furthermore, an occlusion-aware updating strategy is utilized for appearance models, which can easily increase the foreground learning rate. As a result, our method is more suitable for the video sequences with occlusion. Experimental results highlight excellent performance on challenging synthetic and real-world sequences as compared with the state-of-the-art approaches.

Heterogeneous intensity distribution inside the prostate gland, significant variations in prostate shape, size, inter dataset contrast variations, and imaging artifacts like shadow regions and speckle in Trans Rectal Ultrasound (TRUS) images challenge computer aided automatic or semi-automatic segmentation of the prostate. In this paper, we propose a supervised learning schema based on random forest for automatic initialization and propagation of statistical shape and appearance model. Parametric representation of the statistical model of shape and appearance is derived from principal component analysis (PCA) of the probability distribution inside the prostate and PCA of the contour landmarks obtained from the training images. Unlike traditional statistical models of shape and intensity priors, the appearance model in this paper is derived from the posterior probabilities obtained from random forest classification. This probabilistic information is then used for the initialization and propagation of the statistical model. The proposed method achieves mean Dice Similarity Coefficient (DSC) value of 0.96±0.01, with a mean segmentation time of 0.67±0.02 seconds when validated with 24 images from 6 datasets with considerable shape, size, and intensity variations, in a leave-one-patient-out validation framework. The model achieves statistically significant t-test p-value<0.0001 in mean DSC and mean mean absolute distance (MAD) values compared to traditional statistical models of shape and intensity priors.

This paper is concerned with the detection of moving objects using a pan-tilt camera and a background subtraction algorithm. Traditionally, motion compensation is performed on the current image to align its pixels with their background models in previous frames. Pixel misalignment however can occur during motion compensation. Although this problem can be alleviated by using pixel motion such as the optic flow, motion information itself can be inaccurate and, together with pixel misalignment, contributes to false positive foreground detection. In this paper, we exploit the fact that pixel misalignment and inaccurate optic flow tend not to occur simultaneously for a pixel. Consequently, we can substantially improve the performance of the background subtraction algorithm by evaluating the marginal statistical models of appearance and motion separately — rather than jointly — in classifying whether a pixel is foreground. We will use experiments to validate our approach and establish its superiority to other competing algorithms in the literature.

This paper examines the process real-world SMT manufacturers are employing to evaluate, select and implement automated optical inspection (AOI) systems for post-placement/pre-reflow inspection during the PCB assembly process. A case study from a prominent European SMT manufacturer describes the process used to evaluate a post-placement/pre-reflow AOI system. Traditional, algorithm-based AOI techniques are briefly described and evaluated, along with their strengths and limitations. A new AOI methodology, called statistical appearance modeling, and a stereo-vision inspection machine are briefly described and compared to traditional AOI systems. Examples of payback and ROI calculations and analysis are highlighted. The system requirements are specified and the actual results of the AOI performance trial are summarized. Other relevant topics include analyzing GR&amp;R, interpreting SPC data, and implementing post/placement/pre-reflow AOI.

Active Appearance Models (AAMs) are statistical models of shape and appearance widely used in computer vision to detect landmarks on objects like faces. Fitting an AAM to a new image can be formulated as a non-linear least-squares problem which is typically solved using iterative methods. Owing to its efficiency, Gauss-Newton optimization has been the standard choice over more sophisticated approaches like Newton. In this paper, we show that the AAM problem has structure which can be used to solve efficiently the original Newton problem without any approximations. We then make connections to the original Gauss-Newton algorithm and study experimentally the effect of the additional terms introduced by the Newton formulation on both fitting accuracy and convergence. Based on our derivations, we also propose a combined Newton and Gauss-Newton method which achieves promising fitting and convergence performance. Our findings are validated on two challenging in-the-wild data sets.

Automated staging of knee osteoarthritis severity using deep neural networks

A novel method for the automated extraction of important geometrical parameters of the pelvis and hips from APR images is presented. The shape and intensity variations in APR images are encompassed by the statistical shape and appearance models built from a set of training images for each of the three anatomies, i.e., pelvis, right and left hip, separately. The identification of the pelvis and hips is defined as a flexible object recognition problem, which is solved by generating anatomically plausible object instances and matching them to the APR image. The criterion function minimizes the resulting match error and considers the object topology. The obtained flexible object defines the positions of anatomical landmarks, which are further used to calculate the hip joint contact stress. A leave-one-out test was used to evaluate the performance of the proposed method on a set of 26 APR images. The results show the method is able to properly treat image variations and can reliably and accurately identify anatomies in the image and extract the anatomical landmarks needed in the hip joint contact stress calculation.

We present a novel local region approach for statistically characterizing appearance in the context of medical image segmentation via deformable models. Our appearance model reflects the inhomogeneity of tissue mixtures around the exterior of the object of interest by determining mixture-consistent local region types relative to the object boundary. The region types are formed by clustering local regional image descriptors. We partition the object boundary according to region type and apply principal component analysis on the cluster populations to acquire a statistical model of object appearance that accounts for local variability in the object exterior. We present results using this approach to segment bladders and prostates in CT in the context of day-to-day adaptive radiotherapy for prostate cancer. Results show improved fits versus those obtained with a previously developed method

Prostate volume estimation from segmentation of transrectal ultrasound (TRUS) images aids in diagnosis and treatment of prostate hypertrophy and cancer. Computer-aided accurate and computationally efficient prostate segmentation in TRUS images is a challenging task, owing to low signal-to-noise ratio, speckle noise, calcifications, and heterogeneous intensity distribution in the prostate region. A multi-resolution framework using texture features in a parametric deformable statistical model of shape and appearance was developed to segment the prostate. Local phase information of log-Gabor quadrature filter extracted texture of the prostate region in TRUS images. Large bandwidth of log-Gabor filter ensures easy estimation of local orientations, and zero response for a constant signal provides invariance to gray level shift. This aids in enhanced representation of the underlying texture information of the prostate unaffected by speckle noise and imaging artifacts. The parametric model of the propagating contour is derived from principal component analysis of prior shape and texture information of the prostate from the training data. The parameters were modified using prior knowledge of the optimization space to achieve segmentation. The proposed method achieves a mean Dice similarity coefficient value of 0.95 ± 0.02 and mean absolute distance of 1.26 ± 0.51 millimeter when validated with 24 TRUS images of 6 data sets in a leave-one-patient-out validation framework. The proposed method for prostate TRUS image segmentation is computationally efficient and provides accurate prostate segmentations in the presence of intensity heterogeneities and imaging artifacts.

Statistical shape and appearance models are proving valuable tools for medical image analysis. In this paper we outline the methodology and highlight important recent developments. In particular, we consider an extension to images of differing modality, and describe an optimisation approach to establishing correspondences across a training set of shapes.