Bernard Fertil

SHAPE AND TEXTURE INDEXES APPLICATION TO CELL NUCLEI CLASSIFICATION

This paper describes the sequence of construction of a cell nuclei classification model by the analysis, the characterization and the classification of shape and texture. We describe first the elaboration of dedicated shape indexes and second the construction of the associated classification submodel. Then we present a new method of texture characterization, based on the construction and the analysis of statistical matrices encoding the texture. The various characterization techniques developed in this paper are systematically compared to previous approaches. In particular, we paid special attention to the results obtained by a versatile classification method using a large range of descriptors dedicated to the characterization of shapes and textures. Finally, the last classifier built with our methods achieved 88% of classification out of the 94% possible.

RankVisu: Mapping from the neighborhood network

Most multidimensional scaling methods focus on the preservation of dissimilarities to map high dimensional items in a low-dimensional space. However, the mapping function usually does not consider the preservation of small dissimilarities as important, since the cost is small with respect to the preservation of large dissimilarities. As a consequence, an items neighborhoods may be sacrificed for the benefit of the overall mapping. We have subsequently designed a mapping method devoted to the preservation of neighborhood ranks rather than their dissimilarities: RankVisu. A mapping of data is obtained in which neighborhood ranks are as close as possible according to the original space. A comparison with both metric and non-metric MDS highlights the pros (in particular, cluster enhancement) and cons of RankVisu.

People re-identification across multiple non-overlapping cameras system by appearance classification and silhouette part segmentation

In this paper, we present a new person re-identification method based on appearance classification and silhouette part segmentation. In crowded areas, heads are considered as most apparent parts, hence the typical advantage of using the skeleton graph for the head detection and location of people after partial occlusion. The appearance classification consists in characterizing the appearance of a person into two classes, the frontal and the back appearance, using head detector and the orthogonal iteration algorithm for head pose estimation. The silhouette part segmentation divides the silhouette into three horizontal parts, ideally corresponding to head, torso and legs using skeleton graph and head detector. Our approach is robust to real world situations, in particular to variations in scales, human pose, illumination and clothes appearance changes. It also allows to reduce the confusion cases among people appearance and the amount of falsely matches.

Ugly Duckling Sign as a Major Factor of Efficiency in Melanoma Detection

Importance Understanding the contribution of the ugly duckling sign (a nevus that is obviously different from the others in a given individual) in intrapatient comparative analysis (IPCA) of nevi may help improve the detection of melanoma. Objectives To assess the agreement of dermatologists on identification of the ugly duckling sign and estimate the contribution of IPCA to the diagnosis of melanoma. Design, Setting, and Participants The same 2089 digital images of the nevi of a sample of 80 patients (mean age, 42 years [range, 19-80 years]; 33 men and 47 women), as well as 766 dermoscopic images from a subset of 30 patients (mean age, 40 years [range, 21-75 years]; 12 men and 18 women), were randomly presented to the same 9 dermatologists for blinded assessment from September 22, 2011, to April 1, 2013. The first experiment was designed to mimic an IPCA situation, with images of all nevi of each patient shown to the dermatologists, who were asked to identify ugly duckling nevi (UDN). The second experiment was designed to mimic a lesion-focused analysis to identify morphologically suspicious nevi. Data analysis was conducted from November 1, 2012, to June 1, 2013. Main Outcomes and Measures Number of nevi labeled UDN and morphologically suspicious nevi, specificity of lesion-focused analysis and IPCA, and number of nevi identified for biopsy. Results Of the 2089 clinical images of nevi from 80 patients (median number of nevi per patient, 26 [range, 8-81]) and 766 dermoscopic images (median number of nevi per patient, 19 [range, 8-81]), all melanomas were labeled UDN and as morphologically suspicious nevi by the 9 dermatologists. The median number of UDN detected per patient was 0.8 among the clinical images of nevi (mean, 1.0; range, 0.48-2.03) and 1.26 among the dermoscopic images (mean, 1.4; range, 1.00-2.06). The propensity to consider more or fewer nevi as having ugly duckling signs was independent of the presentation (clinical or dermoscopic). The agreement among the dermatologists regarding UDN was lower with dermoscopic images (mean pairwise agreement, 0.53 for clinical images and 0.50 for dermoscopic images). The specificity of IPCA was 0.96 for clinical images and 0.95 for dermoscopic images vs 0.88 and 0.85, respectively, for lesion-focused analysis. When both IPCA and lesion-focused analyses were used, the number of nevi considered for biopsy was reduced by a factor of 6.9 compared with lesion-focused analysis alone. Conclusions and Relevance Intrapatient comparative analysis is of major importance to the effectiveness of the diagnosis of melanoma. Introducing IPCA using the ugly duckling sign in computer-assisted diagnosis systems would be expected to improve performance.

Person re-identification using appearance classification

In this paper, we present a person re-identification method based on appearance classification. It consists a human silhouette comparison by characterizing and classification of a persons appearance (the front and the back appearance) using the geometric distance between the detected head of person and the camera. The combination of head detector with an orthogonal iteration algorithm to help head pose estimation and appearance classification is the novelty of our work. In this way, the is achieved robustness against viewpoint, illumination and clothes appearance changes. Our approach uses matching of interest-points descriptors based on fast cross-bin metric. The approach applies to situations where the number of people varies continuously, considering multiple images for each individual.

A regression-based method for the prediction of the indecisiveness degree through eye movement patterns

The development of eye-tracking-based methods to describe a persons indecisiveness is not commonly explored, even though research has shown that indecisiveness is involved in many unwanted cognitive states, such as a reduction in self-confidence during the decision-making process, doubts about past decisions, reconsidering, trepidation, distractibility, procrastination, neuroticism and even revenge. The purpose of our work is to propose a predictive model of a subjects degree of indecisiveness. To reach this goal, we first need to extract statistically relevant. Using eye-tracking methodology, we build a list of patterns that best distinguish decisive people from indecisive people; this segmentation is made according to the state of the art. The final list of eye-tracking patterns is also coherent with the state of art. A comparison between Multiple Linear Regression (MLR) and Support Vector Regression (SVR) is made so as to select the best predictive model.

Head detection based on skeleton graph method for counting people in crowded environments

Abstract. We describe a method for detecting heads in order to count people in crowded environments using a single camera. The main difference between this method and traditional ones consists of adapting skeleton graph analysis techniques for distinguishing individuals in crowded environments. First, a graph skeleton is calculated for each selected blob in a scene after having performed motion estimation. Then, the structural property of each blob is explored to detect possible heads in order to estimate the number of people. Each detected head in the skeleton silhouette is identified as being in an independent or partial occlusion state and is updated during a tracking process. Finally, the results of our experiments are presented to demonstrate the robustness of our method.

Texture descriptors based on adaptive neighborhoods for classification of pigmented skin lesions

Abstract. Different texture descriptors are proposed for the automatic classification of skin lesions from dermoscopic images. They are based on color texture analysis obtained from (1) color mathematical morphology (MM) and Kohonen self-organizing maps (SOMs) or (2) local binary patterns (LBPs), computed with the use of local adaptive neighborhoods of the image. Neither of these two approaches needs a previous segmentation process. In the first proposed descriptor, the adaptive neighborhoods are used as structuring elements to carry out adaptive MM operations which are further combined by using Kohonen SOM; this has been compared with a nonadaptive version. In the second one, the adaptive neighborhoods enable geometrical feature maps to be defined, from which LBP histograms are computed. This has also been compared with a classical LBP approach. A receiver operating characteristics analysis of the experimental results shows that the adaptive neighborhood-based LBP approach yields the best results. It outperforms the nonadaptive versions of the proposed descriptors and the dermatologists’ visual predictions.

Structural analysis of fMRI data: a surface-based framework for multi-subject studies.

We present a method for fMRI data group analysis that makes the link between two distinct frameworks: surface-based techniques, which process data in the domain defined by the surface of the cortex, and structural techniques, which use object-based representations of the data as opposed to voxel-based ones. This work is a natural surface-based extension of the volume-based structural approach presented in a previous paper. A multi-scale surface-based representation of individual activation maps is first computed for each subject. Then the inter-subject matching and the activation detection decision are performed jointly by optimization of a Markovian model. Finally, a significance measure is computed in a non-parametric way for the results, in order to assess their relevance and control the risk of type I error. The method is applied on simulated and real data and the results are compared to those produced by standard analyses. The surface-based structural analysis is shown to be particularly robust to inter-subject spatial variability and to produce relevant results with good specificity and sensitivity. We also demonstrate the advantages of the surface-based approach by comparing with the results of a 3D structural analysis.

Identification of purchasing scenarios through eye-tracking features

Eye-tracking-based methods are generating a growing interest in marketing research. Nevertheless, most of the studies are focusing on intention, emotion or the evaluation of the products by the customer. The work that is presented here investigates two of the main purchasing scenarios: the routine purchasing act and the impulse purchasing act. The purpose is to propose a predictive model that best distinguishes the first scenario from the second scenario. To reach this goal, we extract statistically relevant eye-tracking descriptors. We use a supervised learning algorithm, Support Vector Machines (SVM), to build the model and reach performances of 82.5% of good identification.