2 code implementations • 26 Mar 2024 • Antoine Théberge, Maxime Descoteaux, Pierre-Marc Jodoin
Reinforcement learning (RL)-based tractography is a competitive alternative to machine learning and classical tractography algorithms due to its high anatomical accuracy obtained without the need for any annotated data.
no code implementations • 15 Jan 2024 • Nathan Painchaud, Pierre-Yves Courand, Pierre-Marc Jodoin, Nicolas Duchateau, Olivier Bernard
Deep learning now enables automatic and robust extraction of cardiac function descriptors from echocardiographic sequences, such as ejection fraction or strain.
no code implementations • 11 Jul 2023 • Daniel Jörgens, Pierre-Marc Jodoin, Maxime Descoteaux, Rodrigo Moreno
Their outputs are combined to obtain the classification labels for the streamlines.
no code implementations • 23 May 2023 • Bach Kim, Jose Dolz, Pierre-Marc Jodoin, Christian Desrosiers
Our system has two components: 1) a segmentation network on the server side which processes the image mixture, and 2) a segmentation unmixing network which recovers the correct segmentation map from the segmentation mixture.
1 code implementation • 15 May 2023 • Antoine Théberge, Christian Desrosiers, Maxime Descoteaux, Pierre-Marc Jodoin
Recently, deep reinforcement learning (RL) has been proposed to learn the tractography procedure and train agents to reconstruct the structure of the white matter without manually curated reference streamlines.
1 code implementation • 3 May 2023 • Hang Jung Ling, Nathan Painchaud, Pierre-Yves Courand, Pierre-Marc Jodoin, Damien Garcia, Olivier Bernard
Deep learning-based methods have spearheaded the automatic analysis of echocardiographic images, taking advantage of the publication of multiple open access datasets annotated by experts (CAMUS being one of the largest public databases).
no code implementations • 30 Nov 2022 • Félix Dumais, Jon Haitz Legarreta, Carl Lemaire, Philippe Poulin, François Rheault, Laurent Petit, Muhamed Barakovic, Stefano Magon, Maxime Descoteaux, Pierre-Marc Jodoin
Our proposed method improves bundle segmentation coverage by recovering hard-to-track bundles with generative sampling through the latent space seeding of the subject bundle and the atlas bundle.
no code implementations • 23 Nov 2022 • Audrey Duran, Gaspard Dussert, Olivier Rouvière, Tristan Jaouen, Pierre-Marc Jodoin, Carole Lartizien
In this work, we propose a novel end-to-end multi-class network that jointly segments the prostate gland and cancer lesions with GS group grading.
no code implementations • 15 Jun 2022 • Thierry Judge, Olivier Bernard, Mihaela Porumb, Agis Chartsias, Arian Beqiri, Pierre-Marc Jodoin
For this reason, we propose CRISP a ContRastive Image Segmentation for uncertainty Prediction method.
no code implementations • 22 Apr 2022 • Jon Haitz Legarreta, Laurent Petit, Pierre-Marc Jodoin, Maxime Descoteaux
GESTA is thus a novel deep generative bundle tractography method that can be used to improve the tractography reconstruction of the white matter.
1 code implementation • 3 Dec 2021 • Nathan Painchaud, Nicolas Duchateau, Olivier Bernard, Pierre-Marc Jodoin
In this paper, we propose a framework to learn the 2D+time apical long-axis cardiac shape such that the segmented sequences can benefit from temporal and anatomical consistency constraints.
no code implementations • 9 Aug 2021 • Benoit Anctil-Robitaille, Antoine Théberge, Pierre-Marc Jodoin, Maxime Descoteaux, Christian Desrosiers, Hervé Lombaert
The physical and clinical constraints surrounding diffusion-weighted imaging (DWI) often limit the spatial resolution of the produced images to voxels up to 8 times larger than those of T1w images.
no code implementations • 23 Jul 2021 • Youssef Skandarani, Pierre-Marc Jodoin, Alain Lalande
Results reveal that generalization performances of a segmentation neural network trained on non-expert groundtruth data is, to all practical purposes, as good as on expert groundtruth data, in particular when the non-expert gets a decent level of training, highlighting an opportunity for the efficient and cheap creation of annotations for cardiac datasets.
no code implementations • 11 May 2021 • Youssef Skandarani, Pierre-Marc Jodoin, Alain Lalande
The top-performing GANs are capable of generating realistic-looking medical images by FID standards that can fool trained experts in a visual Turing test and comply to some metrics.
Ranked #3 on Medical Image Generation on ACDC
1 code implementation • 2 Dec 2020 • Marco Armenta, Thierry Judge, Nathan Painchaud, Youssef Skandarani, Carl Lemaire, Gabriel Gibeau Sanchez, Philippe Spino, Pierre-Marc Jodoin
In this paper, we explore a process called neural teleportation, a mathematical consequence of applying quiver representation theory to neural networks.
no code implementations • 25 Nov 2020 • Bach Ngoc Kim, Jose Dolz, Christian Desrosiers, Pierre-Marc Jodoin
Results show that the segmentation accuracy of our method is similar to a system trained on non-encoded images, while considerably reducing the ability to recover subject identity.
no code implementations • 7 Oct 2020 • Jon Haitz Legarreta, Laurent Petit, François Rheault, Guillaume Theaud, Carl Lemaire, Maxime Descoteaux, Pierre-Marc Jodoin
Current brain white matter fiber tracking techniques show a number of problems, including: generating large proportions of streamlines that do not accurately describe the underlying anatomy; extracting streamlines that are not supported by the underlying diffusion signal; and under-representing some fiber populations, among others.
no code implementations • 23 Jul 2020 • Marco Antonio Armenta, Pierre-Marc Jodoin
In this work, we show that neural networks can be represented via the mathematical theory of quiver representations.
1 code implementation • 15 Jun 2020 • Nathan Painchaud, Youssef Skandarani, Thierry Judge, Olivier Bernard, Alain Lalande, Pierre-Marc Jodoin
In this paper, we present a framework for producing cardiac image segmentation maps that are guaranteed to respect pre-defined anatomical criteria, while remaining within the inter-expert variability.
no code implementations • MIDL 2019 • Youssef Skandarani, Nathan Painchaud, Pierre-Marc Jodoin, Alain Lalande
On one side of our model is a Variational Autoencoder (VAE) trained to learn the latent representations of cardiac shapes.
no code implementations • 4 Apr 2020 • Sarah Leclerc, Erik Smistad, Andreas Østvik, Frederic Cervenansky, Florian Espinosa, Torvald Espeland, Erik Andreas Rye Berg, Thomas Grenier, Carole Lartizien, Pierre-Marc Jodoin, Lasse Lovstakken, Olivier Bernard
Results obtained on a large open access dataset show that our method outperforms the current best performing deep learning solution and achieved an overall segmentation accuracy lower than the intra-observer variability for the epicardial border (i. e. on average a mean absolute error of 1. 5mm and a Hausdorff distance of 5. 1mm) with 11% of outliers.
no code implementations • MIDL 2019 • Audrey Duran, Pierre-Marc Jodoin, Carole Lartizien
Performance of our model was compared to a U-Net baseline model to assess the impact of the self attention module on PCa detection.
no code implementations • 9 Sep 2019 • Bach Ngoc Kim, Jose Dolz, Pierre-Marc Jodoin, Christian Desrosiers
Our novel architecture is composed of three components: 1) an encoder network which removes identity-specific features from input medical images, 2) a discriminator network that attempts to identify the subject from the encoded images, 3) a medical image analysis network which analyzes the content of the encoded images (segmentation in our case).
no code implementations • 16 Aug 2019 • Sarah Leclerc, Erik Smistad, João Pedrosa, Andreas Østvik, Frederic Cervenansky, Florian Espinosa, Torvald Espeland, Erik Andreas Rye Berg, Pierre-Marc Jodoin, Thomas Grenier, Carole Lartizien, Jan D'hooge, Lasse Lovstakken, Olivier Bernard
Delineation of the cardiac structures from 2D echocardiographic images is a common clinical task to establish a diagnosis.
1 code implementation • 5 Jul 2019 • Nathan Painchaud, Youssef Skandarani, Thierry Judge, Olivier Bernard, Alain Lalande, Pierre-Marc Jodoin
In this paper, we propose a cardiac MRI segmentation method which always produces anatomically plausible results.
1 code implementation • CVPR 2019 • Frederic Branchaud-Charron, Andrew Achkar, Pierre-Marc Jodoin
In this paper, we propose a new measure to gauge the complexity of image classification problems.
no code implementations • 14 Feb 2019 • Philippe Poulin, Daniel Jörgens, Pierre-Marc Jodoin, Maxime Descoteaux
Supervised machine learning (ML) algorithms have recently been proposed as an alternative to traditional tractography methods in order to address some of their weaknesses.
no code implementations • CVPR 2019 • Carl Lemaire, Andrew Achkar, Pierre-Marc Jodoin
Pruning methods have shown to be effective at reducing the size of deep neural networks while keeping accuracy almost intact.
2 code implementations • CVPR 2017 • Zhiming Luo, Akshaya Mishra, Andrew Achkar, Justin Eichel, Shaozi Li, Pierre-Marc Jodoin
Saliency detection aims to highlight the most relevant objects in an image.
Ranked #2 on RGB Salient Object Detection on UCF
no code implementations • 24 May 2017 • Clement Zotti, Zhiming Luo, Alain Lalande, Olivier Humbert, Pierre-Marc Jodoin
In this paper, we propose a fully automatic MRI cardiac segmentation method based on a novel deep convolutional neural network (CNN) designed for the 2017 ACDC MICCAI challenge.
no code implementations • 18 Jul 2016 • Mohammad Havaei, Nicolas Guizard, Hugo Larochelle, Pierre-Marc Jodoin
In this chapter, we provide a survey of CNN methods applied to medical imaging with a focus on brain pathology segmentation.
no code implementations • 5 Oct 2015 • Mohammad Havaei, Hugo Larochelle, Philippe Poulin, Pierre-Marc Jodoin
Purpose: In this paper, we investigate a framework for interactive brain tumor segmentation which, at its core, treats the problem of interactive brain tumor segmentation as a machine learning problem.
14 code implementations • 13 May 2015 • Mohammad Havaei, Axel Davy, David Warde-Farley, Antoine Biard, Aaron Courville, Yoshua Bengio, Chris Pal, Pierre-Marc Jodoin, Hugo Larochelle
Finally, we explore a cascade architecture in which the output of a basic CNN is treated as an additional source of information for a subsequent CNN.
Ranked #1 on Brain Tumor Segmentation on BRATS-2013 leaderboard
no code implementations • 1 May 2014 • Greg Castanon, Mohamed Elgharib, Venkatesh Saligrama, Pierre-Marc Jodoin
We present a content-based retrieval method for long surveillance videos both for wide-area (Airborne) as well as near-field imagery (CCTV).