Publications

Les publications sont affichées dans la langue dans laquelle elles ont été soumises.

Publications

Les publications sont affichées dans la langue dans laquelle elles ont été soumises.

Mai 2019 | Artificial Intelligence Endpoint

Learning to Learn with Conditional Class Dependencies

Publication: International Conference on Learning Representations.

Xiang Jiang, Mohammad Havaei, Farshid Varno, Gabriel Chartrand, Nicolas Chapados, Stan Matwin.

Neural networks can learn to extract statistical properties from data, but they seldom make use of structured information from the label space to help representation learning. For example “cat” and “dog” are closer than “cat” and “truck”. Although some label structure can implicitly be obtained when training on huge amounts of data, in a few-shot learning context where little data is available, making explicit use of the label structure can inform the model to reshape the representation space to reflect a global sense of class dependencies. We propose a meta-learning framework, Conditional class-Aware Meta-Learning (CAML), that conditionally transforms feature representations based on a metric space that is trained to capture inter-class dependencies. This enables a conditional modulation of the feature representations of the base-learner to impose regularities informed by the label space. Experiments show that the conditional transformation in CAML leads to more disentangled representations and achieves competitive results on the miniImageNet benchmark.

Juillet 2018 | Artificial Intelligence Endpoint

Adversarial Mixture Model

Andrew Jesson, Cecile Low-Kam, Tanya Nair, Florian Soudan, Florent Chandelier, Nicolas Chapados.

The Adversarially Learned Mixture Model (AMM) is a generative model for unsupervised or semi-supervised data clustering. The AMM is the first adversarially optimized method to model the conditional dependence between inferred continuous and categorical latent variables. Experiments on the MNIST and SVHN datasets show that the AMM allows for semantic separation of complex data when little or no labeled data is available. The AMM achieves a state-of-the-art unsupervised clustering error rate of 2.86% on the MNIST dataset. A semi-supervised extension of the AMM yields competitive results on the SVHN dataset.

Juin 2018 | Artificial Intelligence Endpoint

On the Importance of Attention in Meta-Learning for Few-Shot Text Classification

Xiang Jiang, Mohammad Havaei, Gabriel Chartrand, Hassan Chouaib, Thomas Vincent, Andrew Jesson, Nicolas Chapados, Stan Matwin.

Current deep learning based text classification methods are limited by their ability to achieve fast learning and generalization when the data is scarce. We address this problem by integrating a meta-learning procedure that uses the knowledge learned across many tasks as an inductive bias towards better natural language understanding. Based on the Model-Agnostic Meta-Learning framework (MAML), we introduce the Attentive Task-Agnostic Meta-Learning (ATAML) algorithm for text classification. The essential difference between MAML and ATAML is in the separation of task-agnostic representation learning and task-specific attentive adaptation. The proposed ATAML is designed to encourage task-agnostic representation learning by way of task-agnostic parameterization and facilitate task-specific adaptation via attention mechanisms. We provide evidence to show that the attention mechanism in ATAML has a synergistic effect on learning performance. In comparisons with models trained from random initialization, pretrained models and meta trained MAML, our proposed ATAML method generalizes better on single-label and multi-label classification tasks in miniRCV1 and miniReuters-21578 datasets.

Avril 2018 | Artificial Intelligence Endpoint

Iteratively unveiling new regions of interest in Deep Learning models

Publication: Submitted to International Conference on Medical Imaging with Deep Learning

Florian Bordes, Tess Berthier, Lisa Di Jorio, Pascal Vincent, Yoshua Bengio

Recent advance of deep learning has been transforming the landscape in many domain, including health care. However, understanding the predictions of a deep network remains a challenge, which is especially sensitive in health care domains as interpretability is key. Techniques that rely on saliency maps -highlighting the region of an image that influence the classifier’s decision the most- are often used for that purpose. However, gradients fluctuation make saliency maps noisy ant thus difficult to interpret at a human level. Moreover, models tend to focus on one particular influential region of interest (ROI) in the image, even though other regions might be relevant for the decision. We propose a new framework that refines those saliency maps to generate segmentation masks over the ROI on the initial image. In a second contribution, we propose to apply those masks over the original inputs, then evaluate our classifier on the masked inputs to identify previously unidentified ROI. This iterative procedure allows us to emphasize new region of interests by extracting meaningful information from the saliency maps.

Février 2018 | Position paper

Role of artificial intelligence in the care of patients
with non-small cell lung cancer

Publication: European Journal of Clinical Investigation

Rabbani M, Kanevsky J, Kafi K, Chandelier F, Giles FJ

Recent studies support the use of computer‐aided systems and the use of radiomic features to help diagnose lung cancer earlier. Other studies have looked at machine learning (ML) methods that offer prognostic tools to doctors and help them in choosing personalized treatment options for their patients based on molecular, genetics and histological features. Combining artificial intelligence approaches into health care may serve as a beneficial tool for patients with NSCLC, and this review outlines these benefits and current shortcomings throughout the continuum of care.

2018 | Artificial Intelligence Endpoint

Deep Learning for Automated Segmentation of Colorectal Cancer Liver Metastases on Computed Tomography

Publication: Submitted to Radiological Society of North America 2018 Scientific Assembly and Annual Meeting

Milena Cerny, Eugene Vorontsov, Philippe Régnier, Lisa Di Jorio, Chris Pal, Réal Lapointe, Franck Vanderbroucke, Simon Turcotte, Samuel Kadoury, An Tang

Novembre 2017 | Position Paper

Deep Learning: A Primer for Radiologists

Publication: Radiographics

Gabriel Chartrand, PhD, Phillip M. Cheng, MD, MS, Eugene Vorontsov, BASc Eng Sci, Michal Drozdzal, PhD, Simon Turcotte, MD, MSc, Christopher J. Pal, PhD, Samuel Kadoury, PhD, An Tang, MD, MSc.

Deep learning is a class of machine learning methods that are gaining success and attracting interest in many domains, including computer vision, speech recognition, natural language processing, and playing games. Deep learning methods produce a mapping from raw inputs to desired outputs (eg, image classes). Unlike traditional machine learning methods, which require hand-engineered feature extraction from inputs, deep learning methods learn these features directly from data. With the advent of large datasets and increased computing power, these methods can produce models with exceptional performance. These models are multilayer artificial neural networks, loosely inspired by biologic neural systems. Weighted connections between nodes (neurons) in the network are iteratively adjusted based on example pairs of inputs and target outputs by back-propagating a corrective error signal through the network. For computer vision tasks, convolutional neural networks (CNNs) have proven to be effective. Recently, several clinical applications of CNNs have been proposed and studied in radiology for classification, detection, and segmentation tasks. This article reviews the key concepts of deep learning for clinical radiologists, discusses technical requirements, describes emerging applications in clinical radiology, and outlines limitations and future directions in this field. Radiologists should become familiar with the principles and potential applications of deep learning in medical imaging.

 

Novembre 2017 | Clinical Endpoint

Deep Learning: A Primer for Radiologists

Publication: Radiographics

Gabriel Chartrand, PhD, Phillip M. Cheng, MD, MS, Eugene Vorontsov, BASc Eng Sci, Michal Drozdzal, PhD, Simon Turcotte, MD, MSc, Christopher J. Pal, PhD, Samuel Kadoury, PhD, An Tang, MD, MSc.

Deep learning is a class of machine learning methods that are gaining success and attracting interest in many domains, including computer vision, speech recognition, natural language processing, and playing games. Deep learning methods produce a mapping from raw inputs to desired outputs (eg, image classes). Unlike traditional machine learning methods, which require hand-engineered feature extraction from inputs, deep learning methods learn these features directly from data. With the advent of large datasets and increased computing power, these methods can produce models with exceptional performance. These models are multilayer artificial neural networks, loosely inspired by biologic neural systems. Weighted connections between nodes (neurons) in the network are iteratively adjusted based on example pairs of inputs and target outputs by back-propagating a corrective error signal through the network. For computer vision tasks, convolutional neural networks (CNNs) have proven to be effective. Recently, several clinical applications of CNNs have been proposed and studied in radiology for classification, detection, and segmentation tasks. This article reviews the key concepts of deep learning for clinical radiologists, discusses technical requirements, describes emerging applications in clinical radiology, and outlines limitations and future directions in this field. Radiologists should become familiar with the principles and potential applications of deep learning in medical imaging.

 

Octobre 2017 | Clinical Endpoint

Real-time differentiation of adenomatous and hyperplastic diminutive colorectal polyps during analysis of unaltered videos of standard colonoscopy using a deep learning mode

Publication: Gut

Michael F Byrne, Nicolas Chapados, Florian Soudan, Clemens Oertel, Milagros Linares Pérez, Raymond Kelly, Nadeem Iqbal, Florent Chandelier, Douglas K Rex

 

We developed an artificial intelligence (AI) model for real-time assessment of endoscopic video images of colorectal polyps. A deep convolutional neural network model was used. Only narrow band imaging video frames were used, split equally between relevant multiclasses. Unaltered videos from routine exams not specifically designed or adapted for AI classification were used to train and validate the model. The model was tested on a separate series of 125 videos of consecutively encountered diminutive polyps that were proven to be adenomas or hyperplastic polyps.

Soumis le 6 octobre 2017 | Artificial Intelligence Endpoint

Learnable Explicit Density for Continuous Latent Space and Variational Inference.

Chin-Wei Huang, Ahmed Touati, Laurent Dinh, Michal Drozdzal, Mohammad Havaei, Laurent Charlin, Aaron Courville.

In this paper, we study two aspects of the variational autoencoder (VAE): the prior distribution over the latent variables and its corresponding posterior. First, we decompose the learning of VAEs into layerwise density estimation, and argue that having a flexible prior is beneficial to both sample generation and inference. Second, we analyze the family of inverse autoregressive flows (inverse AF) and show that with further improvement, inverse AF could be used as universal approximation to any complicated posterior. Our analysis results in a unified approach to parameterizing a VAE, without the need to restrict ourselves to use factorial Gaussians in the latent real space.

Septembre 2017 | Artificial Intelligence Endpoint

CASED: Curriculum Adaptive Sampling for Extreme Data Imbalance.

Publication:  International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 639-646. Springer, Cham.

Andrew Jesson, Nicolas Guizard, Sina Hamidi Ghalehjegh, Damien Goblot, Florian Soudan, and Nicolas Chapados.

We introduce CASED, a novel curriculum sampling algorithm that facilitates the optimization of deep learning segmentation or detection models on data sets with extreme class imbalance. We evaluate the CASED learning framework on the task of lung nodule detection in chest CT. In contrast to two-stage solutions wherein nodule candidates are first proposed by a segmentation model and refined by a second detection stage, CASED improves the training of deep nodule segmentation models (e.g. UNet) to the point where state of the art results are achieved using only a trivial detection stage. CASED improves the optimization of deep segmentation models by allowing them to first learn how to distinguish nodules from their immediate surroundings, while continuously adding a greater proportion of difficult-to-classify global context, until uniformly sampling from the empirical data distribution. Using CASED during training yields a minimalist proposal to the lung nodule detection problem that tops the LUNA16 nodule detection benchmark with an average sensitivity score of 88.35%. Furthermore, we find that models trained using CASED are robust to nodule annotation quality by showing that comparable results can be achieved when only a point and radius for each ground truth nodule are provided during training. Finally, the CASED learning framework makes no assumptions with regard to imaging modality or segmentation target and so should generalize to other medical imaging problems where class imbalance is a persistent problem.

Août 2017 | Artificial Intelligence Endpoint

Liver lesion segmentation informed by joint liver segmentation.

Eugene Vorontsov, An Tang, Chris Pal, Samuel Kadouryblot, Florian Soudan, and Nicolas Chapados.

 

We propose a model for the joint segmentation of the liver and liver lesions in computed tomography (CT) volumes. We build the model from two fully convolutional networks, connected in tandem and trained together end-to-end. We evaluate our approach on the 2017 MICCAI Liver Tumour Segmentation Challenge, attaining competitive liver and liver lesion detection and segmentation scores across a wide range of metrics. Unlike other top performing methods, our model output post-processing is trivial, we do not use data external to the challenge, and we propose a simple single-stage model that is trained end-to-end. However, our method nearly matches the top lesion segmentation performance and achieves the second highest precision for lesion detection while maintaining high recall.

Août 2017 | Artificial Intelligence Endpoint

Image Segmentation by Iterative Inference from Conditional Score Estimation.

Adriana Romero, Michal Drozdzal, Akram Erraqabi, Simon Jégou, Yoshua Bengio.

Inspired by the combination of feedforward and iterative computations in the virtual cortex, and taking advantage of the ability of denoising autoencoders to estimate the score of a joint distribution, we propose a novel approach to iterative inference for capturing and exploiting the complex joint distribution of output variables conditioned on some input variables. This approach is applied to image pixel-wise segmentation, with the estimated conditional score used to perform gradient ascent towards a mode of the estimated conditional distribution. This extends previous work on score estimation by denoising autoencoders to the case of a conditional distribution, with a novel use of a corrupted feedforward predictor replacing Gaussian corruption. An advantage of this approach over more classical ways to perform iterative inference for structured outputs, like conditional random fields (CRFs), is that it is not any more necessary to define an explicit energy function linking the output variables. To keep computations tractable, such energy function parametrizations are typically fairly constrained, involving only a few neighbors of each of the output variables in each clique. We experimentally find that the proposed iterative inference from conditional score estimation by conditional denoising autoencoders performs better than comparable models based on CRFs or those not using any explicit modeling of the conditional joint distribution of outputs.

Février 2017 | Artificial Intelligence Endpoint

Learning Normalized Inputs for Iterative Estimation in Medical Image Segmentation.

Drozdzal, Michal, Gabriel Chartrand, Eugene Vorontsov, Lisa Di Jorio, An Tang, Adriana Romero, Yoshua Bengio, Chris Pal, and Samuel Kadoury.

 

In this paper, we introduce a simple, yet powerful pipeline for medical image segmentation that combines Fully Convolutional Networks (FCNs) with Fully Convolutional Residual Networks (FC-ResNets). We propose and examine a design that takes particular advantage of recent advances in the understanding of both Convolutional Neural Networks as well as ResNets. Our approach focuses upon the importance of a trainable pre-processing when using FC-ResNets and we show that a low-capacity FCN model can serve as a pre-processor to normalize medical input data. In our image segmentation pipeline, we use FCNs to obtain normalized images, which are then iteratively refined by means of a FC-ResNet to generate a segmentation prediction. As in other fully convolutional approaches, our pipeline can be used off-the-shelf on different image modalities. We show that using this pipeline, we exhibit state-of-the-art performance on the challenging Electron Microscopy benchmark, when compared to other 2D methods. We improve segmentation results on CT images of liver lesions, when contrasting with standard FCN methods. Moreover, when applying our 2D pipeline on a challenging 3D MRI prostate segmentation challenge we reach results that are competitive even when compared to 3D methods. The obtained results illustrate the strong potential and versatility of the pipeline by achieving highly accurate results on multi-modality images from different anatomical regions and organs.

En révision | Artificial Intelligence Endpoint

Learning latent state representations from partial observations.

Michal Drozdzal*, Mohammad Havaei*, Chin-Wei Huang, Nicolas Chapados, Laurent Charlin, Aaron Courville.

*equal contribution

Décembre 2016 | Artificial Intelligence Endpoint

A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images.

Vázquez, David, Jorge Bernal, F. Javier Sánchez, Gloria Fernández-Esparrach, Antonio M. López, Adriana Romero, Michal Drozdzal, and Aaron Courville.

 

Colorectal cancer (CRC) is the third cause of cancer death worldwide. Currently, the standard approach to reduce CRC-related mortality is to perform regular screening in search for polyps and colonoscopy is the screening tool of choice. The main limitations of this screening procedure are polyp miss-rate and inability to perform visual assessment of polyp malignancy. These drawbacks can be reduced by designing Decision Support Systems (DSS) aiming to help clinicians in the different stages of the procedure by providing endoluminal scene segmentation. Thus, in this paper, we introduce an extended benchmark of colonoscopy image, with the hope of establishing a new strong benchmark for colonoscopy image analysis research. We provide new baselines on this dataset by training standard fully convolutional networks (FCN) for semantic segmentation and significantly outperforming, without any further post-processing, prior results in endoluminal scene segmentation.

Novembre 2016 | Position Paper

Deep Learning: A Primer for Radiologists

Publication: Radiological Society of North America 2016 Scientific Assembly and Annual Meeting. Chicago, Il

Chartrand, Gabriel, Eugene Vorontsov, Mathieu Flamand, Simon Turcotte, Christopher Pal, Samuel Kadoury, and An Tang.

Novembre 2016 | Artificial Intelligence Endpoint

The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation.

Jégou, Simon, Michal Drozdzal, David Vazquez, Adriana Romero, and Yoshua Bengio.

 

State-of-the-art approaches for semantic image segmentation are built on Convolutional Neural Networks (CNNs). The typical segmentation architecture is composed of (a) a downsampling path responsible for extracting coarse semantic features, followed by (b) an upsampling path trained to recover the input image resolution at the output of the model and, optionally, (c) a post-processing module (e.g. Conditional Random Fields) to refine the model predictions. Recently, a new CNN architecture, Densely Connected Convolutional Networks (DenseNets), has shown excellent results on image classification tasks. The idea of DenseNets is based on the observation that if each layer is directly connected to every other layer in a feed-forward fashion then the network will be more accurate and easier to train.
In this paper, we extend DenseNets to deal with the problem of semantic segmentation. We achieve state-of-the-art results on urban scene benchmark datasets such as CamVid and Gatech, without any further post-processing module nor pretraining. Moreover, due to smart construction of the model, our approach has much less parameters than currently published best entries for these datasets.

Novembre 2016 | Artificial Intelligence Endpoint

Automated Segmentation of Liver Metastases with Deep Convolutional Neural Networks.

Radiological Society of North America 2016 Scientific Assembly and Annual Meeting. Chicago, Il, 2016.

Vorontsov, Eugene, Gabriel Chartrand, Olina Dagher, Vi Thuy Tran, Mathieu Flamand, Aline Khatchikian, Amine Smouk, et al.

Août 2016 | Artificial Intelligence Endpoint

The Importance of Skip Connections in Biomedical Image Segmentation.

Drozdzal, Michal, Eugene Vorontsov, Gabriel Chartrand, Samuel Kadoury, and Chris Pal.

In this paper, we study the influence of both long and short skip connections on Fully Convolutional Networks (FCN) for biomedical image segmentation. In standard FCNs, only long skip connections are used to skip features from the contracting path to the expanding path in order to recover spatial information lost during downsampling. We extend FCNs by adding short skip connections, that are similar to the ones introduced in residual networks, in order to build very deep FCNs (of hundreds of layers). A review of the gradient flow confirms that for a very deep FCN it is beneficial to have both long and short skip connections. Finally, we show that a very deep FCN can achieve near-to-state-of-the-art results on the EM dataset without any further post-processing.

Juillet 2016 | Artificial Intelligence Endpoint

HeMIS: Hetero-Modal Image Segmentation.

MICCAI 2016. Lecture Notes in Computer Science, Vol 9901, 469–77. Springer, Cham, 2016.

Havaei, Mohammad, Nicolas Guizard, Nicolas Chapados, and Yoshua Bengio.

We introduce a deep learning image segmentation framework that is extremely robust to missing imaging modalities. Instead of attempting to impute or synthesize missing data, the proposed approach learns, for each modality, an embedding of the input image into a single latent vector space for which arithmetic operations (such as taking the mean) are well defined. Points in that space, which are averaged over modalities available at inference time, can then be further processed to yield the desired segmentation. As such, any combinatorial subset of available modalities can be provided as input, without having to learn a combinatorial number of imputation models. Evaluated on two neurological MRI datasets (brain tumors and MS lesions), the approach yields state-of-the-art segmentation results when provided with all modalities; moreover, its performance degrades remarkably gracefully when modalities are removed, significantly more so than alternative mean-filling or other synthesis approaches.

Juillet 2016 | Artificial Intelligence Endpoint

Deep Learning Trends for Focal Brain Pathology Segmentation in MRI.

MICCAI 2016. Lecture Notes in Computer Science, Vol 9901, 469–77. Springer, Cham, 2016.

Havaei, Mohammad, Nicolas Guizard, Nicolas Chapados, and Yoshua Bengio.

 

Segmentation of focal (localized) brain pathologies such as brain tumors and brain lesions caused by multiple sclerosis and ischemic strokes are necessary for medical diagnosis, surgical planning and disease development as well as other applications such as tractography. Over the years, attempts have been made to automate this process for both clinical and research reasons. In this regard, machine learning methods have long been a focus of attention. Over the past two years, the medical imaging field has seen a rise in the use of a particular branch of machine learning commonly known as deep learning. In the non-medical computer vision world, deep learning based methods have obtained state-of-the-art results on many datasets. Recent studies in computer aided diagnostics have shown deep learning methods (and especially convolutional neural networks – CNN) to yield promising results. In this chapter, we provide a survey of CNN methods applied to medical imaging with a focus on brain pathology segmentation. In particular, we discuss their characteristic peculiarities and their specific configuration and adjustments that are best suited to segment medical images. We also underline the intrinsic differences deep learning methods have with other machine learning methods.

2016 | Clinical Endpoint

Deep Learning: A Primer for Radiologists.

Publication: Radiological Society of North America 2016 Scientific Assembly and Annual Meeting. Chicago, Il,

Chartrand, Gabriel, Eugene Vorontsov, Mathieu Flamand, Simon Turcotte, Christopher Pal, Samuel Kadoury, and An Tang.

2016 | Clinical Endpoint

Artificial Intelligence (Ai) In Endoscopy–Deep Learning For Optical Biopsy Of Colorectal Polyps In Real-Time On Unaltered Endoscopic Videos.

Publication: United European Gastroenterology Journal, 4:A155, 2016. doi:10.1177/2050640616663688.

Byrne, Michael F., Douglas K. Rex, Nicolas Chapados, Florian Soudan, Clemens Oertel, Milagros Linares Perez, R. Kelly, N. Iqbal, and Florent Chandelier.

We trained our model on 223 videos and validated on 40 videos in white light (WL) and narrow band imaging (NBI). We tested the model on 125 videos of colorectal polyps (≤ 5mm) under NBI, using Olympus 190 series scopes. Pathology of resected polyps was the reference standard. We used a Deep Learning Artificial Intelligence model with a proprietary deep convolutional neural network (DCNN) for NICE type 1&2 differentiation. The model operated at the individual frame level, without prior segmentation.