Publications

Synthetic data as an enabler for machine learning applications in medicine
Jean-Francois Rajotte
Robert Bergen
David L Buckeridge
Khaled El Emam
Raymond Ng
Elissa Strome
2022-09-30
iScience (publié)
doi.org
The use of artificial intelligence and virtual reality in doctor-patient risk communication: A scoping review.
Ryan Antel
S. A. Rahimi
Elena Guadagno
Jason M. Harley
Dan Poenaru
2022-09-30
Patient Education and Counseling (publié)
doi.org
Evolution of cell size control is canalized towards adders or sizers by cell cycle structure and selective pressures
Felix Proulx-Giraldeau
Jan M Skotheim
Paul François
Cell size is controlled to be within a specific range to support physiological function. To control their size, cells use diverse mechanisms… (voir plus) ranging from ‘sizers’, in which differences in cell size are compensated for in a single cell division cycle, to ‘adders’, in which a constant amount of cell growth occurs in each cell cycle. This diversity raises the question why a particular cell would implement one rather than another mechanism? To address this question, we performed a series of simulations evolving cell size control networks. The size control mechanism that evolved was influenced by both cell cycle structure and specific selection pressures. Moreover, evolved networks recapitulated known size control properties of naturally occurring networks. If the mechanism is based on a G1 size control and an S/G2/M timer, as found for budding yeast and some human cells, adders likely evolve. But, if the G1 phase is significantly longer than the S/G2/M phase, as is often the case in mammalian cells in vivo, sizers become more likely. Sizers also evolve when the cell cycle structure is inverted so that G1 is a timer, while S/G2/M performs size control, as is the case for the fission yeast S. pombe. For some size control networks, cell size consistently decreases in each cycle until a burst of cell cycle inhibitor drives an extended G1 phase much like the cell division cycle of the green algae Chlamydomonas. That these size control networks evolved such self-organized criticality shows how the evolution of complex systems can drive the emergence of critical processes.
2022-09-29
eLife (publié)
doi.org
Learning Robust Kernel Ensembles with Kernel Average Pooling
Pouya Bashivan
Adam Ibrahim
Amirozhan Dehghani
Yifei Ren
2022-09-29
ArXiv (prépublication)
doi.org
openreview.net
SPeCiaL: Self-Supervised Pretraining for Continual Learning
Lucas Caccia
Joelle Pineau
2022-09-27
Continual Semi-Supervised Learning (publié)
doi.org
arxiv.org
From analytic to synthetic-organizational pluralisms: A pluralistic enactive psychiatry
Christophe Gauld
Kristopher Nielsen
Manon Job
Hugo Bottemanne
Guillaume Dumas
Reliance on sole reductionism, whether explanatory, methodological or ontological, is difficult to support in clinical psychiatry. Rather, p… (voir plus)sychiatry is challenged by a plurality of approaches. There exist multiple legitimate ways of understanding human functionality and disorder, i.e., different systems of representation, different tools, different methodologies and objectives. Pluralistic frameworks have been presented through which the multiplicity of approaches in psychiatry can be understood. In parallel of these frameworks, an enactive approach for psychiatry has been proposed. In this paper, we consider the relationships between the different kinds of pluralistic frameworks and this enactive approach for psychiatry. We compare the enactive approach in psychiatry with wider analytical forms of pluralism. On one side, the enactive framework anchored both in cognitive sciences, theory of dynamic systems, systems biology, and phenomenology, has recently been proposed as an answer to the challenge of an integrative psychiatry. On the other side, two forms of explanatory pluralisms can be described: a non-integrative pluralism and an integrative pluralism. The first is tolerant, it examines the coexistence of different potentially incompatible or untranslatable systems in the scientific or clinical landscape. The second is integrative and proposes to bring together the different levels of understanding and systems of representations. We propose that enactivism is inherently a form of integrative pluralism, but it is at the same time a component of the general framework of explanatory pluralism, composed of a set of so-called analytical approaches. A significant number of mental health professionals are already accepting the variety of clinical and scientific approaches. In this way, a rigorous understanding of the theoretical positioning of psychiatric actors seems necessary to promote quality clinical practice. The study of entanglements between an analytical pluralism and a synthetic-organizational enactivist pluralism could prove fruitful.
2022-09-26
Frontiers in Psychiatry (publié)
doi.org
FedShuffle: Recipes for Better Use of Local Work in Federated Learning
Samuel Horváth
Maziar Sanjabi
Lin Xiao
Peter Richtárik
Michael G. Rabbat
The practice of applying several local updates before aggregation across clients has been empirically shown to be a successful approach to o… (voir plus)vercoming the communication bottleneck in Federated Learning (FL). Such methods are usually implemented by having clients perform one or more epochs of local training per round while randomly reshuffling their finite dataset in each epoch. Data imbalance, where clients have different numbers of local training samples, is ubiquitous in FL applications, resulting in different clients performing different numbers of local updates in each round. In this work, we propose a general recipe, FedShuffle, that better utilizes the local updates in FL, especially in this regime encompassing random reshuffling and heterogeneity. FedShuffle is the first local update method with theoretical convergence guarantees that incorporates random reshuffling, data imbalance, and client sampling — features that are essential in large-scale cross-device FL. We present a comprehensive theoretical analysis of FedShuffle and show, both theoretically and empirically, that it does not suffer from the objective function mismatch that is present in FL methods that assume homogeneous updates in heterogeneous FL setups, such as FedAvg (McMahan et al., 2017). In addition, by combining the ingredients above, FedShuffle improves upon FedNova (Wang et al., 2020), which was previously proposed to solve this mismatch. Similar to Mime (Karimireddy et al., 2020), we show that FedShuffle with momentum variance reduction (Cutkosky & Orabona, 2019) improves upon non-local methods under a Hessian similarity assumption.
2022-09-25
TMLR (accepté)
doi.org
openreview.net
Tackling bias in AI health datasets through the STANDING Together initiative
Shaswath Ganapathi
Johannes Palmer
J. Alderman
Melanie Calvert
Cyrus Espinoza
Jacqui Gath
Marzyeh Ghassemi
Katherine Heller
Francis McKay
Alan Karthikesalingam
S. Kuku
Maxine E. Mackintosh
Sinduja Manohar
Bilal Mateen
Rubeta Matin
Melissa D. McCradden
Lauren Oakden-Rayner
Johan Ordish
Russell Pearson
S. Pfohl … (voir 8 de plus)
Negar Rostamzadeh
Elizabeth Sapey
Neil J. Sebire
Viknesh Sounderajah
Charlotte Summers
Darren E. Treanor
Alastair Denniston
Xiaoxuan Liu
2022-09-25
Nature Network Boston (publié)
doi.org
The 5-year longitudinal diagnostic profile and health services utilization of patients treated with electroconvulsive therapy in Quebec: a population-based study
Simon Lafrenière
Fatemeh Gholi-Zadeh-Kharrat
Caroline Sirois
Victoria Massamba
Louis Rochette
Camille Brousseau-Paradis
Simon Patry
Christian Gagné
Morgane Lemasson
Geneviève Gariépy
Chantal Mérette
Elham Rahme
Alain Lesage
2022-09-25
Social Psychiatry and Psychiatric Epidemiology (publié)
doi.org
OSSEM: One-Shot Speaker Adaptive Speech Enhancement Using Meta Learning
Cheng Yu
Szu-wei Fu
Tsun-An Hsieh
Yu Tsao
Mirco Ravanelli
Although deep learning (DL) has achieved notable progress in speech enhancement (SE), further research is still required for a DL-based SE s… (voir plus)ystem to adapt effectively and efficiently to particular speakers. In this study, we propose a novel meta-learning-based speaker-adaptive SE approach (called OSSEM) that aims to achieve SE model adaptation in a one-shot manner. OSSEM consists of a modified transformer SE network and a speaker-specific masking (SSM) network. In practice, the SSM network takes an enrolled speaker embedding extracted using ECAPA-TDNN to adjust the input noisy feature through masking. To evaluate OSSEM, we designed a modified Voice Bank-DEMAND dataset, in which one utterance from the testing set was used for model adaptation, and the remaining utterances were used for testing the performance. Moreover, we set restrictions allowing the enhancement process to be conducted in real time, and thus designed OSSEM to be a causal SE system. Experimental results first show that OSSEM can effectively adapt a pretrained SE model to a particular speaker with only one utterance, thus yielding improved SE results. Meanwhile, OSSEM exhibits a competitive performance compared to state-of-the-art causal SE systems.
2022-09-17
Interspeech 2022 (publié)
doi.org
arxiv.org
SoundChoice: Grapheme-to-Phoneme Models with Semantic Disambiguation
Artem Ploujnikov
Mirco Ravanelli
End-to-end speech synthesis models directly convert the input characters into an audio representation (e.g., spectrograms). Despite their im… (voir plus)pressive performance, such models have difficulty disambiguating the pronunciations of identically spelled words. To mitigate this issue, a separate Grapheme-to-Phoneme (G2P) model can be employed to convert the characters into phonemes before synthesizing the audio. This paper proposes SoundChoice, a novel G2P architecture that processes entire sentences rather than operating at the word level. The proposed architecture takes advantage of a weighted homograph loss (that improves disambiguation), exploits curriculum learning (that gradually switches from word-level to sentence-level G2P), and integrates word embeddings from BERT (for further performance improvement). Moreover, the model inherits the best practices in speech recognition, including multi-task learning with Connectionist Temporal Classification (CTC) and beam search with an embedded language model. As a result, SoundChoice achieves a Phoneme Error Rate (PER) of 2.65% on whole-sentence transcription using data from LibriSpeech and Wikipedia. Index Terms grapheme-to-phoneme, speech synthesis, text-tospeech, phonetics, pronunciation, disambiguation.
2022-09-17
Interspeech 2022 (publié)
doi.org
arxiv.org
Intervertebral Disc Labeling With Learning Shape Information, A Look Once Approach
Reza Azad
Moein Heidari
Julien Cohen-Adad
Ehsan Adeli
Dorit Merhof
Accurate and automatic segmentation of intervertebral discs from medical images is a critical task for the assessment of spine-related disea… (voir plus)ses such as osteoporosis, vertebral fractures, and intervertebral disc herniation. To date, various approaches have been developed in the literature which routinely relies on detecting the discs as the primary step. A disadvantage of many cohort studies is that the localization algorithm also yields false-positive detections. In this study, we aim to alleviate this problem by proposing a novel U-Net-based structure to predict a set of candidates for intervertebral disc locations. In our design, we integrate the image shape information (image gradients) to encourage the model to learn rich and generic geometrical information. This additional signal guides the model to selectively emphasize the contextual representation and suppress the less discriminative features. On the post-processing side, to further decrease the false positive rate, we propose a permutation invariant 'look once' model, which accelerates the candidate recovery procedure. In comparison with previous studies, our proposed approach does not need to perform the selection in an iterative fashion. The proposed method was evaluated on the spine generic public multi-center dataset and demonstrated superior performance compared to previous work. We have provided the implementation code in https://github.com/rezazad68/intervertebral-lookonce
2022-09-15
Predictive Intelligence in Medicine (publié)
doi.org
arxiv.org