Non-resonant inelastic X-ray scattering for discrimination of pigments.

Inelastic X-ray scattering (IXS) spectroscopy has been used in many fields of solid-state physics and theoretical chemistry as an accurate and quantitative probe of elementary excitations. We show that non-resonant IXS spectra in the energy loss range below 100 eV exhibit a strong contrast across a wide range of commercially available pigments, opening new routes for their discrimination. These signatures combine plasmonic transitions, collective excitations and low energy absorption edges. We have performed IXS to discriminate different artists' pigments within complex mixtures and to quantitatively determine rutile and anatase polymorphs of TiO2. The integration of experimental data on pigment powders with suitable ab initio simulations shows a precise fit of the spectroscopic data both in the position of the resonances and in their relative intensity.

Exploiting Patch Sizes and Resolutions for Multi-Scale Deep Learning in Mammogram Image Classification.

Recent progress in deep learning (DL) has revived the interest on DL-based computer aided detection or diagnosis (CAD) systems for breast cancer screening. Patch-based approaches are one of the main state-of-the-art techniques for 2D mammogram image classification, but they are intrinsically limited by the choice of patch size, as there is no unique patch size that is adapted to all lesion sizes. In addition, the impact of input image resolution on performance is not yet fully understood. In this work, we study the impact of patch size and image resolution on the classifier performance for 2D mammograms. To leverage the advantages of different patch sizes and resolutions, a multi patch-size classifier and a multi-resolution classifier are proposed. These new architectures perform multi-scale classification by combining different patch sizes and input image resolutions. The AUC is increased by 3% on the public CBIS-DDSM dataset and by 5% on an internal dataset. Compared with a baseline single patch size and single resolution classifier, our multi-scale classifier reaches an AUC of 0.809 and 0.722 in each dataset.

A 3D Mathematical Breast Texture Model With Parameters Automatically Inferred From Clinical Breast CT Images.

A numerical realistic 3D anthropomorphic breast model is useful for evaluating breast imaging applications. A method is proposed to model small and medium-scale fibroglandular and intra-glandular adipose tissues observed in the center part of clinical breast CT images. The method builds upon a previously proposed model formulated as stochastic geometric processes with mathematically tractable parameters. In this work, the medium-scale parameters were automatically and objectively inferred from breast CT images. We hypothesized that a set of random ellipsoids exhibiting cluster interaction is representative to model the medium-scale intra-glandular adipose compartments. The ellipsoids were reconstructed using a multiple birth, death and shift algorithm. Then, a Matérn cluster process was used to fit the reconstructed ellipsoid centers. Finally, distributions of the ellipsoid shapes and orientations were estimated using maximum likelihood estimators. Feasibility was demonstrated on 16 volumes of interests (VOI). To assess the realism of the 3D breast texture model, ß and LFE metrics computed in simulated projection images of simulated texture realizations and clinical images were compared. Visual realism was illustrated. For 12 out of 16 VOIs, our hypothesis on clustering interaction process is confirmed. The average ß values from simulated texture images (3.7 to 4.2) of the 12 different VOIs are higher than the average ß value from 2D clinical images (2.87). LFE of simulated texture images and clinical mammograms are similar. Compared to our previous model, whereby simulation parameters were based upon empirical observations, our inference method substantially augments the ability to generate textures with higher visual realism and larger morphological variety.

Bayesian technique for image classifying registration.

In this paper, we address a complex image registration issue arising while the dependencies between intensities of images to be registered are not spatially homogeneous. Such a situation is frequently encountered in medical imaging when a pathology present in one of the images modifies locally intensity dependencies observed on normal tissues. Usual image registration models, which are based on a single global intensity similarity criterion, fail to register such images, as they are blind to local deviations of intensity dependencies. Such a limitation is also encountered in contrast-enhanced images where there exist multiple pixel classes having different properties of contrast agent absorption. In this paper, we propose a new model in which the similarity criterion is adapted locally to images by classification of image intensity dependencies. Defined in a Bayesian framework, the similarity criterion is a mixture of probability distributions describing dependencies on two classes. The model also includes a class map which locates pixels of the two classes and weighs the two mixture components. The registration problem is formulated both as an energy minimization problem and as a maximum a posteriori estimation problem. It is solved using a gradient descent algorithm. In the problem formulation and resolution, the image deformation and the class map are estimated simultaneously, leading to an original combination of registration and classification that we call image classifying registration. Whenever sufficient information about class location is available in applications, the registration can also be performed on its own by fixing a given class map. Finally, we illustrate the interest of our model on two real applications from medical imaging: template-based segmentation of contrast-enhanced images and lesion detection in mammograms. We also conduct an evaluation of our model on simulated medical data and show its ability to take into account spatial variations of intensity dependencies while keeping a good registration accuracy.

A classifying registration technique for the estimation of enhancement curves of DCE-CT scan sequences.

In this paper, we propose a new technique for the estimation of contrast enhancement curves of Dynamic Contrast-Enhanced sequences, which takes the most from the interdependence between this estimation problem and the registration problem raised by possible movements occurring in sequences. The technique solves the estimation and registration problems simultaneously in an iterative way. However, unlike previous techniques, a pixel classification scheme is included within the estimation so as to compute enhancement curves on pixel classes instead of single pixels. The classification scheme is designed using a descendant hierarchical approach. Due to this tree approach, the number of classes is set automatically and the whole technique is entirely unsupervised. Moreover, some specific prior information about the shape of enhancement curves are included in the splitting and pruning steps of the classification scheme. Such an information ensures that created classes include pixels having homogeneous and relevant enhancement properties. The technique is applied to DET-CT scan sequences and evaluated using ground truth data. Results show that classifications are anatomically sound and that contrast enhancements are accurately estimated from sequences.

A nonparametric approach for histogram segmentation.

In this work, we propose a method to segment a 1-D histogram without a priori assumptions about the underlying density function. Our approach considers a rigorous definition of an admissible segmentation, avoiding over and under segmentation problems. A fast algorithm leading to such a segmentation is proposed. The approach is tested both with synthetic and real data. An application to the segmentation of written documents is also presented. We shall see that this application requires the detection of very small histogram modes, which can be accurately detected with the proposed method.

Computational gestalts and perception thresholds.

In 1923, Max Wertheimer proposed a research programme and method in visual perception. He conjectured the existence of a small set of geometric grouping laws governing the perceptual synthesis of phenomenal objects, or "gestalt" from the atomic retina input. In this paper, we review this set of geometric grouping laws, using the works of Metzger, Kanizsa and their schools. In continuation, we explain why the Gestalt theory research programme can be translated into a Computer Vision programme. This translation is not straightforward, since Gestalt theory never addressed two fundamental matters: image sampling and image information measurements. Using these advances, we shall show that gestalt grouping laws can be translated into quantitative laws allowing the automatic computation of gestalts in digital images. From the psychophysical viewpoint, a main issue is raised: the computer vision gestalt detection methods deliver predictable perception thresholds. Thus, we are set in a position where we can build artificial images and check whether some kind of agreement can be found between the computationally predicted thresholds and the psychophysical ones. We describe and discuss two preliminary sets of experiments, where we compared the gestalt detection performance of several subjects with the predictable detection curve. In our opinion, the results of this experimental comparison support the idea of a much more systematic interaction between computational predictions in Computer Vision and psychophysical experiments.

Dequantizing image orientation.

We address the problem of computing a local orientation map in a digital image. We show that standard image gray level quantization causes a strong bias in the repartition of orientations, hindering any accurate geometric analysis of the image. In continuation, a simple dequantization algorithm is proposed, which maintains all of the image information and transforms the quantization noise in a nearby Gaussian white noise (we actually prove that only Gaussian noise can maintain isotropy of orientations). Mathematical arguments are used to show that this results in the restoration of a high quality image isotropy. In contrast with other classical methods, it turns out that this property can be obtained without smoothing the image or increasing the signal-to-noise ratio (SNR). As an application, it is shown in the experimental section that, thanks to this dequantization of orientations, such geometric algorithms as the detection of nonlocal alignments can be performed efficiently. We also point out similar improvements of orientation quality when our dequantization method is applied to aliased images.