Program Committee

SPECIAL SESSION GSI 2023: GEOMETRIC FEATURES EXTRACTION IN MEDICAL IMAGING

International Conference on Geometric Science of Information GSI 2023, editors :Frank Nielsen and Frédéric Barbaresco

We organize a special session with P. Clarysse (DR CNRS, CREATIS) on "geometric features extraction in medical imaging" and we participate to the program committee of the conference

Please find below the list of the invited papers, and use this link to access to the book edited by Springer with the full papers in pdf.

Diffeomorphic ICP registration for single and multiple point sets, Adrien Wohrer 

Abstract: We propose a generalization of the ICP registration algorithm which accommodates diffeomorphic mappings in the LDDMM framework. The algorithm is formulated as a well-posed probabilistic inference, and requires to solve a novel variation of LDDMM landmark registration with an additional term involving the Jacobian of the mapping. The algorithm can easily be generalized to construct a diffeomorphic, statistical atlas of multiple point sets. The method is successfully validated on a first set of synthetic data. 

Chan-Vese Attention U-Net: An attention mechanism for robust segmentation, Nicolas Makaroff and Laurent D. Cohen 

Abstract: When studying the results of a segmentation algorithm using convolutional neural networks, one wonders how robust the results are. This leads to questioning the possibility of using such an algorithm in applications where there is little room for doubt. We propose in this paper a new attention gate based on the use of ChanVese energy minimization to control more precisely the segmentation masks given by a standard CNN architecture such as the U-Net model. This mechanism allows to obtain a constraint on the segmentation based on the resolution of a PDE. The study of the results allows us to observe the spatial information retained by the neural network on the region of interest and obtains competitive results on the binary segmentation. We illustrate the efficiency of this approach for medical image segmentation on a database of brain images. 

Using a Riemannian elastic metric for statistical analysis of tumor cell shape heterogeneity, Wanxin Li, Ashok Prasad, Nina Miolane, and Khanh Dao Duc 

Abstract: We examine how a specific instance of the elastic metric, the Square Root Velocity (SRV) metric, can be used to study and compare cellular morphologies from the contours they form on planar surfaces. We process a dataset of images from osteocarcoma (bone cancer) cells that includes different treatments known to affect the cell morphology, and perform a comparative statistical analysis between the linear and SRV metrics. Our study indicates superior performance of the SRV at capturing the cell shape heterogeneity, with a better separation between different cell groups when comparing their distance to their mean shape, as well as a better low dimensional representation when comparing stress statistics. Therefore, our study suggests the use of a Riemannian metric, such as the SRV as a potential tool to enhance morphological discrimination for large datasets of cancer cell images. 

Perturbation of Fiedler vector: interest for graph measures and shape analysis, Julien Lefevre, Justine Fraize, and David Germanaud 

Abstract: In this paper we investigate some properties of the Fiedler vector, the so-called first non-trivial eigenvector of the Laplacian matrix of a graph. There are important results about the Fiedler vector to identify spectral cuts in graphs but far less is known about its extreme values and points. We propose a few results and conjectures in this direction. We also bring two concrete contributions, i) by defining a new measure for graphs that can be interpreted in terms of extremality (inverse of centrality), ii) by applying a small perturbation to the Fiedler vector of cerebral shapes such as the corpus callosum to robustify their parameterization. 

PROGRAM COMMITTEE: CAD4MED workshop 2023

In 2023, I was part of the program committee of the 2nd International Workshop on Computer-Aided Diagnosis for Biomedical Applications whose chairman was S. Bricq, Université de Bourgogne, France.

This workshop aims to provide an overview of recent advances in the field of Computer-Aided Diagnosis Systems for Biomedical Applications, covering all the medical signal/image processing workflow, from signal/image acquisition to classification, including disease detection, prediction, and classification.

Link towards the workshop: https://www.sitis-conference.org/2023/contribute/workshops/cad4med/?cmplz-force-reload=1714402511401

2021