DKDenis Kouame

Denis Kouame

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Denis Kouame has a total of 49 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A novel perspective on denoising using quantum localization with application to medical imaging

    by Amirreza Hashemi, Sayantan Dutta, Bertrand Georgeot, Denis Kouamé, Hamid Sabet

    Background noise in many fields, such as medical imaging, poses significant challenges for accurate diagnosis, prompting the development of denoising algorithms. Traditional methodologies, however, often struggle to address the complexities of noisy environments in high dimensional imaging systems. This paper introduces a novel quantum-inspired approach for image denoising, drawing upon principles of quantum and condensed matter physics. Our approach views medical images as amorphous structures akin to those found in condensed matter physics, and we propose an algorithm that incorporates the concept of mode resolved localization directly into the denoising process. Notably, unlike previous studies that considered localization as a hindrance, our approach considers quantum localization as a fundamental component of image reconstruction, which is used to differentiate between noisy and non-noisy modes based on diffusivity and localization measurements. This perspective eliminates the need for hyperparameter tuning, making the proposed method a standalone algorithm that can be implemented with minimal manual intervention and can perform automatic filtering of noise regardless of noise level. Through numerical validation, we showcase the effectiveness of our approach in addressing noise-related challenges in imaging, and especially medical imaging, underscoring its relevance for possible quantum computing applications.


  • Computational Super-Resolution for Ultrasound Localization Microscopy Through Solving an Inverse Problem

    by Vassili Pustovalov, Duong Hung Pham, Corentin Alix, Jean-Pierre Remeniéras, Denis Kouamé

    Ultrasound localization microscopy (ULM) represents a significant advancement over traditional ultrasound (US) imaging, enabling super-resolution (SR) imaging of microvascular structures with unprecedented detail, by using microbubbles (MBs) as highly reflective contrast agents. After injection into the bloodstream, MBs are localized in US images to reconstruct the microvasculature. However, this technique faces a trade-off between MB localization accuracy and acquisition time. Perfusion with low MB concentrations reduces signal overlap and achieves high localization accuracy but requires extended acquisition times. Conversely, higher MB concentrations shorten acquisition times but increase signal overlap, limiting localization precision. Traditionally, ULM consists of five main steps: tissue filtering, MB detection, MB super-localization, tracking, and rendering. In this study, we present a novel approach that combines a robust principal component analysis (RPCA) with a computational SR technique, replacing the first three steps of ULM with a single process based on solving a SR inverse problem. This method isolates MB signals from background noise and enhances the localization of overlapping MBs, thereby improving overall ULM performance. Experimental results from simulated and in vivo data demonstrate that our proposed approach increases the SR factor by up to 30% and enhances the contrast ratio (CR) by 3.5 dB. It also produces comparable results across other image quality metrics. These improvements enable denser, higher-contrast vascular images.


  • Automatic Tuning of Denoising Algorithms Parameters Without Ground Truth

    by Arthur Floquet, Sayantan Dutta, Emmanuel Soubies, Duong-Hung Pham, Denis Kouame

    Denoising is omnipresent in image processing. It is usually addressed with algorithms relying on a set of hyperparameters that control the quality of the recovered image. Manual tuning of those parameters can be a daunting task, which calls for the development of automatic tuning methods. Given a denoising algorithm, the best set of parameters is the one that minimizes the error between denoised and ground-truth images. Clearly, this ideal approach is unrealistic, as the ground-truth images are unknown in practice. In this work, we propose unsupervised cost functions—i.e., that only require the noisy image—that allow us to reach this ideal gold standard performance. Specifically, the proposed approach makes it possible to obtain an average PSNR output within less than 1% of the best achievable PSNR.


  • Quantum Algorithm for Signal Denoising

    by Sayantan Dutta, Adrian Basarab, Denis Kouamé, Bertrand Georgeot

    This letter presents a novel quantum algorithm for signal denoising, which performs a thresholding in the frequency domain through amplitude amplification and using an adaptive threshold determined by local mean values. The proposed algorithm is able to process both classical and quantum signals. It is parametrically faster than previous classical and quantum denoising algorithms. Numerical results show that it is efficient at removing noise of both classical and quantum origin, significantly outperforming existing quantum algorithms in this respect, especially in the presence of quantum noise.


  • CX-DaGAN: Domain Adaptation for Pneumonia Diagnosis on a Small Chest X-Ray Dataset

    by Karen Sanchez, Carlos Hinojosa, Henry Arguello, Denis Kouame, Olivier Meyrignac, Adrian Basarab

    Recent advances in deep learning led to several algorithms for the accurate diagnosis of pneumonia from chest X-rays. However, these models require large training medical datasets, which are sparse, isolated, and generally private. Furthermore, these models in medical imaging are known to over-fit to a particular data domain source, i.e., these algorithms do not conserve the same accuracy when tested on a dataset from another medical center, mainly due to image distribution discrepancies. In this work, a domain adaptation and classification technique is proposed to overcome the over-fit challenges on a small dataset. This method uses a private-small dataset (target domain), a public-large labeled dataset from another medical center (source domain), and consists of three steps. First, it performs a data selection of the source domain’s most representative images based on similarity constraints through principal component analysis subspaces. Second, the selected samples from the source domain are fit to the target distribution through an image to image translation based on a cycle-generative adversarial network. Finally, the target train dataset and the adapted images from the source dataset are used within a convolutional neural network to explore different settings to adjust the layers and perform the classification of the target test dataset. It is shown that fine-tuning a few specific layers together with the selected-adapted images increases the sorting accuracy while reducing the trainable parameters. The proposed approach achieved a notable increase in the target dataset’s overall classification accuracy, reaching up to 97.78% compared to 90.03% by standard transfer learning.


  • Open Access

    Single Image Super-Resolution Of Noisy 3d Dental Ct Images Using Tucker Decomposition

    by J. Hatvaniy, J. Michetti, A. Basarab, M. Gyongy, D. Kouame

    Tensor decomposition has proven to be a strong tool in various 3D image processing tasks such as denoising and superresolution. In this context, we recently proposed a computationally effective, canonical polyadic decomposition (CPD) based algorithm for single image super-resolution. In this work, we investigated the added value brought by Tucker decomposition. While CPD allows a joint implementation of the denoising and deconvolution steps, with Tucker decomposition the denoising is followed by the deconvolution. This way the ill-posedness of the deconvolution caused by noise is partially mitigated. The results achieved using the two different tensor decomposition techniques were compared, and the robustness against noise was investigated. For validation, dental images were used. The superiority of the proposed method is shown in terms of peak signal-to-ratio, structural similarity index, the root canal segmentation accuracy, and runtime.


  • Open Access

    Quantum Mechanics-Based Signal and Image Representation: Application to Denoising

    by Sayantan Dutta, Adrian Basarab, Bertrand Georgeot, Denis Kouame

    Decomposition of digital signals and images into other basis or dictionaries than time or space domains is a very common approach in signal and image processing and analysis. Such a decomposition is commonly obtained using fixed transforms (e.g., Fourier or wavelet) or dictionaries learned from example databases or from the signal or image itself. In this work, we investigate in detail a new approach of constructing such a signal or image-dependent bases inspired by quantum mechanics tools, i.e., by considering the signal or image as a potential in the discretized Schroedinger equation. To illustrate the potential of the proposed decomposition, denoising results are reported in the case of Gaussian, Poisson, and speckle noise and compared to the state of the art algorithms based on wavelet shrinkage, total variation regularization or patch-wise sparse coding in learned dictionaries, non-local means image denoising, and graph signal processing.


  • Open Access

    Preconditioned P-ULA for Joint Deconvolution-Segmentation of Ultrasound Images -- Extended Version

    by Corbineau Marie-Caroline, Kouamé Denis, Chouzenoux Emilie, Tourneret Jean-Yves, Pesquet Jean-Christophe

    Joint deconvolution and segmentation of ultrasound images is a challenging problem in medical imaging. By adopting a hierarchical Bayesian model, we propose an accelerated Markov chain Monte Carlo scheme where the tissue reflectivity function is sampled thanks to a recently introduced proximal unadjusted Langevin algorithm. This new approach is combined with a forward-backward step and a preconditioning strategy to accelerate the convergence, and with a method based on the majorization-minimization principle to solve the inner nonconvex minimization problems. As demonstrated in numerical experiments conducted on both simulated and in vivo ultrasound images, the proposed method provides high-quality restoration and segmentation results and is up to six times faster than an existing Hamiltonian Monte Carlo method.