To prospectively validate the ability of customized three-dimensional (3D) software to enable bro... more To prospectively validate the ability of customized three-dimensional (3D) software to enable bronchial tree skeletonization, orthogonal reconstruction of the main bronchial axis, and measurement of cross-sectional wall area (WA) and lumen area (LA) of any visible bronchus on thin-section computed tomographic (CT) images. Institutional review board approval and patient agreement and informed consent were obtained. Software was validated in a phantom that consisted of seven tubes and an excised human lung obtained and used according to institutional guidelines. In vivo validation was performed with multi-detector row CT in six healthy subjects (mean age, 47 years; range, 20-55 years). Intra- and interobserver agreement and reproducibility over time for bronchial tree skeletonization were evaluated with Bland-Altman analysis. Concordance in identifying bronchial generation was assessed with the kappa statistic. WA and LA obtained with the manual method were compared with WA and LA obtained with validated software by means of the Wilcoxon test and Bland-Altman analysis. WA and LA measurements in the phantom were reproducible over multiple sessions (P > .90) and were not significantly different from WA and LA assessed with the manual method (P > .62). WA and LA measurements in the excised lung and the subjects were not different from measurements obtained with the manual method (intraclass correlation coefficient > 0.99). All lobar bronchi and 80.8% of third generation bronchi, 72.5% of fourth generation bronchi, and 37.7% of fifth generation bronchi were identified in vivo. Intra- and interobserver agreement and reproducibility over time for airway skeletonization and concordance in identifying bronchial generation were good to excellent (intraclass correlation coefficient > 0.98, kappa > 0.54, respectively). This method enables accurate and reproducible measurement of WA and LA on reformatted CT sections perpendicular to the main axis of bronchi visible on thin-section CT scans.
Proceedings of the 9th International Conference on Discrete Geometry For Computer Imagery, 2000
In this work we describe a geometric method to smooth the boundary of a discrete 3D object. The m... more In this work we describe a geometric method to smooth the boundary of a discrete 3D object. The method is reversible in the sense that the discrete boundary can be retrieved by digitizing the smoothed one. To this end, we propose a representation of the boundary of a discrete volume that we call Euclidean net and which is a generalization
2009 Seventh International Workshop on Content Based Multimedia Indexing, Jun 1, 2009
Content based image retrieval is a topic which has received a lot of attention and increasing pop... more Content based image retrieval is a topic which has received a lot of attention and increasing popularity due to a wide range of applications. In this paper, we present a similarity measure for CBIR in an industrial context, where the images of a vibration phenomenon are obtained by electronic speckle pattern interferometry. Images obtained have very poor visual characteristics, and
In this paper we are interested in classification of objects in images according to user defined ... more In this paper we are interested in classification of objects in images according to user defined scenarios. We show how the user-defined ontology with a specialisation by a concrete scenario / object of interest allows for an adapted choice of methods and their tuning through the whole framework: se-lection of the area of interest, descriptors choice, classifica-tion of objects. Particular attention here is payed to the classification. We use SVM classifiers for their good capacity of general-isation. We show that in an adapted descriptor space, the choice of a "light" linear kernel together with boosting of classifiers is interesting compared to more complex and com-putationally expensive RBF kernels. The results on real-life images are promising. The paper results from the research we conduct in the framework of X-Media EU-funded Integrated Project.
A method is proposed which computes the visible points of surfaces in a 3-dimensional discrete sp... more A method is proposed which computes the visible points of surfaces in a 3-dimensional discrete space. The occlusion of surface points of an object by other object points is determined by shooting a discrete ray from each surface point towards the center of projection considering the intersection of the ray with other object points. Since the projection of points onto the viewing plane is done by a continuous mapping, additionally to the discrete ray, the location of the continuous projection ray is examined regarding its location to the surface points that are intersected by the discrete ray.
In this work we describe a geometric method to smooth the boundary of a discrete 3D object. The m... more In this work we describe a geometric method to smooth the boundary of a discrete 3D object. The method is reversible in the sense that the discrete boundary can be retrieved by digitizing the smoothed one. To this end, we propose a representation of the boundary of a discrete volume that we call Euclidean net and which is a generalization
We define a new model for describing the resulting color of a stack of 2D objects in 212D renderi... more We define a new model for describing the resulting color of a stack of 2D objects in 212D rendering. We restrict ourselves to objects that transform the color spectrum of incoming light in a linear way. Our model is more accurate than previous ones in that it is a better approximation of the physical behavior of real colored objects. It furthermore has the advantage of being independent of the model used for light. In our model, a graphic object in 212 dimensions is described as a 2 by 2 matrix that combines incoming light from the front and the back to yield resulting light in both directions. We derive an operation that takes the matrices of two graphic objects and combines them into one matrix describing the color behavior of the combined object and we show that this operation is associative. Very accurate results are obtained when using the spectral model of light and distorted but plausible results when working with RGB colors rather than spectra.
ABSTRACT The extraction of a 3D topological map from an Oriented Boundary Graph can be needed to ... more ABSTRACT The extraction of a 3D topological map from an Oriented Boundary Graph can be needed to refine a 3D Split and Merge segmentation using topological information such as the Euler characteristic of regions. A presegmentation could thus be efficiently obtained using a light structuring model, before proceeding to the extraction of more information on some regions of interest. In this paper, we present the topological map extraction algorithm which allows to reconstruct locally a set of regions from the Oriented Boundary Graph. A comparison of the two models is also presented.
5th International Conference on Visual Information Engineering (VIE 2008), 2008
ABSTRACT This article addresses the field of content based image retrieval on large collection of... more ABSTRACT This article addresses the field of content based image retrieval on large collection of images on a query by example basis. We propose a multidimensional organisation of the image database based on several clusterings performed on low level descriptors separately. We show that using this multidimensional index for querying the database outperforms any of the single dimensional query. The other advantage of this method is that it enables to combine several low level descriptors in a single query while keeping the physical meaning of each descriptor in mind. The results on the Amsterdam Library of Object Images are convincing.
To prospectively validate the ability of customized three-dimensional (3D) software to enable bro... more To prospectively validate the ability of customized three-dimensional (3D) software to enable bronchial tree skeletonization, orthogonal reconstruction of the main bronchial axis, and measurement of cross-sectional wall area (WA) and lumen area (LA) of any visible bronchus on thin-section computed tomographic (CT) images. Institutional review board approval and patient agreement and informed consent were obtained. Software was validated in a phantom that consisted of seven tubes and an excised human lung obtained and used according to institutional guidelines. In vivo validation was performed with multi-detector row CT in six healthy subjects (mean age, 47 years; range, 20-55 years). Intra- and interobserver agreement and reproducibility over time for bronchial tree skeletonization were evaluated with Bland-Altman analysis. Concordance in identifying bronchial generation was assessed with the kappa statistic. WA and LA obtained with the manual method were compared with WA and LA obtained with validated software by means of the Wilcoxon test and Bland-Altman analysis. WA and LA measurements in the phantom were reproducible over multiple sessions (P > .90) and were not significantly different from WA and LA assessed with the manual method (P > .62). WA and LA measurements in the excised lung and the subjects were not different from measurements obtained with the manual method (intraclass correlation coefficient > 0.99). All lobar bronchi and 80.8% of third generation bronchi, 72.5% of fourth generation bronchi, and 37.7% of fifth generation bronchi were identified in vivo. Intra- and interobserver agreement and reproducibility over time for airway skeletonization and concordance in identifying bronchial generation were good to excellent (intraclass correlation coefficient > 0.98, kappa > 0.54, respectively). This method enables accurate and reproducible measurement of WA and LA on reformatted CT sections perpendicular to the main axis of bronchi visible on thin-section CT scans.
Proceedings of the 9th International Conference on Discrete Geometry For Computer Imagery, 2000
In this work we describe a geometric method to smooth the boundary of a discrete 3D object. The m... more In this work we describe a geometric method to smooth the boundary of a discrete 3D object. The method is reversible in the sense that the discrete boundary can be retrieved by digitizing the smoothed one. To this end, we propose a representation of the boundary of a discrete volume that we call Euclidean net and which is a generalization
2009 Seventh International Workshop on Content Based Multimedia Indexing, Jun 1, 2009
Content based image retrieval is a topic which has received a lot of attention and increasing pop... more Content based image retrieval is a topic which has received a lot of attention and increasing popularity due to a wide range of applications. In this paper, we present a similarity measure for CBIR in an industrial context, where the images of a vibration phenomenon are obtained by electronic speckle pattern interferometry. Images obtained have very poor visual characteristics, and
In this paper we are interested in classification of objects in images according to user defined ... more In this paper we are interested in classification of objects in images according to user defined scenarios. We show how the user-defined ontology with a specialisation by a concrete scenario / object of interest allows for an adapted choice of methods and their tuning through the whole framework: se-lection of the area of interest, descriptors choice, classifica-tion of objects. Particular attention here is payed to the classification. We use SVM classifiers for their good capacity of general-isation. We show that in an adapted descriptor space, the choice of a "light" linear kernel together with boosting of classifiers is interesting compared to more complex and com-putationally expensive RBF kernels. The results on real-life images are promising. The paper results from the research we conduct in the framework of X-Media EU-funded Integrated Project.
A method is proposed which computes the visible points of surfaces in a 3-dimensional discrete sp... more A method is proposed which computes the visible points of surfaces in a 3-dimensional discrete space. The occlusion of surface points of an object by other object points is determined by shooting a discrete ray from each surface point towards the center of projection considering the intersection of the ray with other object points. Since the projection of points onto the viewing plane is done by a continuous mapping, additionally to the discrete ray, the location of the continuous projection ray is examined regarding its location to the surface points that are intersected by the discrete ray.
In this work we describe a geometric method to smooth the boundary of a discrete 3D object. The m... more In this work we describe a geometric method to smooth the boundary of a discrete 3D object. The method is reversible in the sense that the discrete boundary can be retrieved by digitizing the smoothed one. To this end, we propose a representation of the boundary of a discrete volume that we call Euclidean net and which is a generalization
We define a new model for describing the resulting color of a stack of 2D objects in 212D renderi... more We define a new model for describing the resulting color of a stack of 2D objects in 212D rendering. We restrict ourselves to objects that transform the color spectrum of incoming light in a linear way. Our model is more accurate than previous ones in that it is a better approximation of the physical behavior of real colored objects. It furthermore has the advantage of being independent of the model used for light. In our model, a graphic object in 212 dimensions is described as a 2 by 2 matrix that combines incoming light from the front and the back to yield resulting light in both directions. We derive an operation that takes the matrices of two graphic objects and combines them into one matrix describing the color behavior of the combined object and we show that this operation is associative. Very accurate results are obtained when using the spectral model of light and distorted but plausible results when working with RGB colors rather than spectra.
ABSTRACT The extraction of a 3D topological map from an Oriented Boundary Graph can be needed to ... more ABSTRACT The extraction of a 3D topological map from an Oriented Boundary Graph can be needed to refine a 3D Split and Merge segmentation using topological information such as the Euler characteristic of regions. A presegmentation could thus be efficiently obtained using a light structuring model, before proceeding to the extraction of more information on some regions of interest. In this paper, we present the topological map extraction algorithm which allows to reconstruct locally a set of regions from the Oriented Boundary Graph. A comparison of the two models is also presented.
5th International Conference on Visual Information Engineering (VIE 2008), 2008
ABSTRACT This article addresses the field of content based image retrieval on large collection of... more ABSTRACT This article addresses the field of content based image retrieval on large collection of images on a query by example basis. We propose a multidimensional organisation of the image database based on several clusterings performed on low level descriptors separately. We show that using this multidimensional index for querying the database outperforms any of the single dimensional query. The other advantage of this method is that it enables to combine several low level descriptors in a single query while keeping the physical meaning of each descriptor in mind. The results on the Amsterdam Library of Object Images are convincing.
Uploads
Papers by Achille Braquelaire