Publications

Evaluating the Fairness of Deep Learning Uncertainty Estimates in Medical Image Analysis
Raghav Mehta
Changjian Shui
Tal Arbel
Although deep learning (DL) models have shown great success in many medical image analysis tasks, deployment of the resulting models into r… (voir plus)eal clinical contexts requires: (1) that they exhibit robustness and fairness across different sub-populations, and (2) that the confidence in DL model predictions be accurately expressed in the form of uncertainties. Unfortunately, recent studies have indeed shown significant biases in DL models across demographic subgroups (e.g., race, sex, age) in the context of medical image analysis, indicating a lack of fairness in the models. Although several methods have been proposed in the ML literature to mitigate a lack of fairness in DL models, they focus entirely on the absolute performance between groups without considering their effect on uncertainty estimation. In this work, we present the first exploration of the effect of popular fairness models on overcoming biases across subgroups in medical image analysis in terms of bottom-line performance, and their effects on uncertainty quantification. We perform extensive experiments on three different clinically relevant tasks: (i) skin lesion classification, (ii) brain tumour segmentation, and (iii) Alzheimer's disease clinical score regression. Our results indicate that popular ML methods, such as data-balancing and distributionally robust optimization, succeed in mitigating fairness issues in terms of the model performances for some of the tasks. However, this can come at the cost of poor uncertainty estimates associated with the model predictions. This tradeoff must be mitigated if fairness models are to be adopted in medical image analysis.
2023-04-04
MIDL.io/2023/Conference (poster)
doi.org
openreview.net
Evaluating the Fairness of Deep Learning Uncertainty Estimates in Medical Image Analysis
Raghav Mehta
Changjian Shui
Tal Arbel
2023-04-04
MIDL.io/2023/Conference (poster)
doi.org
openreview.net
Intent-aware Multi-source Contrastive Alignment for Tag-enhanced Recommendation
Haolun Wu
Yingxue Zhang
Chen Ma
Wei Guo
Ruiming Tang
Xue (Steve) Liu
Mark Coates
To offer accurate and diverse recommendation services, recent methods use auxiliary information to foster the learning process of user and i… (voir plus)tem representations. Many state-of-the-art (SOTA) methods fuse different sources of information (user, item, knowledge graph, tags, etc.) into a graph and use Graph Neural Networks (GNNs) to introduce the auxiliary information through the message passing paradigm. In this work, we seek an alternative framework that is light and effective through self-supervised learning across different sources of information, particularly for the commonly accessible item tag information. We use a self-supervision signal to pair users with the auxiliary information (tags) associated with the items they have interacted with before. To achieve the pairing, we create a proxy training task. For a given item, the model predicts which is the correct pairing between the representations obtained from the users that have interacted with this item and the tags assigned to it. This design provides an efficient solution, using the auxiliary information directly to enhance the quality of user and item embeddings. User behavior in recommendation systems is driven by the complex interactions of many factors behind the users’ decision-making processes. To make the pairing process more fine-grained and avoid embedding collapse, we propose a user intent-aware self-supervised pairing process where we split the user embeddings into multiple sub-embedding vectors. Each sub-embedding vector captures a specific user intent via self-supervised alignment with a particular cluster of tags. We integrate our designed framework with various recommendation models, demonstrating its flexibility and compatibility. Through comparison with numerous SOTA methods on seven real-world datasets, we show that our method can achieve better performance while requiring less training time. This indicates the potential of applying our approach on web-scale datasets.
2023-04-03
2023 IEEE 39th International Conference on Data Engineering (ICDE) (publié)
doi.org
arxiv.org
Assessing the inclusion of children’s surgical care in National Surgical, Obstetric and Anaesthesia Plans: a policy content analysis
Sabrina Wimmer
Paul Truche
Elena Guadagno
Emmanuel Ameh
Lubna Samad
Emmanuel Mwenda Malabo Makasa
Sarah Greenberg
John G Meara
Tonnis H van Dijk
Dan Poenaru
Objective While National Surgical, Obstetric and Anaesthesia Plans (NSOAPs) have emerged as a strategy to strengthen and scale up surgical h… (voir plus)ealthcare systems in low/middle-income countries (LMICs), the degree to which children’s surgery is addressed is not well-known. This study aims to assess the inclusion of children’s surgical care among existing NSOAPs, identify practice examples and provide recommendations to guide inclusion of children’s surgical care in future policies. Design We performed two qualitative content analyses to assess the inclusion of children’s surgical care among NSOAPs. We applied a conventional (inductive) content analysis approach to identify themes and patterns, and developed a framework based on the Global Initiative for Children’s Surgery’s Optimal Resources for Children’s Surgery document. We then used this framework to conduct a directed (deductive) content analysis of the NSOAPs of Ethiopia, Nigeria, Rwanda, Senegal, Tanzania and Zambia. Results Our framework for the inclusion of children’s surgical care in NSOAPs included seven domains. We evaluated six NSOAPs with all addressing at least two of the domains. All six NSOAPs addressed ‘human resources and training’ and ‘infrastructure’, four addressed ‘service delivery’, three addressed ‘governance and financing’, two included ‘research, evaluation and quality improvement’, and one NSOAP addressed ‘equipment and supplies’ and ‘advocacy and awareness’. Conclusions Additional focus must be placed on the development of surgical healthcare systems for children in LMICs. This requires a focus on children’s surgical care separate from adult surgical care in the scaling up of surgical healthcare systems, including children-focused needs assessments and the inclusion of children’s surgery providers in the process. This study proposes a framework for evaluating NSOAPs, highlights practice examples and suggests recommendations for the development of future policies.
2023-04-01
BMJ Open (publié)
doi.org
Biomedical discovery through the integrative biomedical knowledge hub (iBKH).
Chang Su
Yu Hou
Manqi Zhou
Suraj Rajendran
Jacqueline R.M. A. Maasch
Zehra Abedi
Haotan Zhang
Zilong Bai
Anthony Cuturrufo
Winston Guo
Fayzan F. Chaudhry
Gregory Ghahramani
Jian Tang
Feixiong Cheng
Yue Li
Rui Zhang
Steven T. DeKosky
Jiang Bian
Fei Wang
2023-04-01
iScience (publié)
doi.org
Can Ensembling Preprocessing Algorithms Lead to Better Machine Learning Fairness?
Khaled Badran
Pierre-Olivier Côté
Amanda Kolopanis
Rached Bouchoucha
Antonio Collante
Diego Elias Costa
Emad Shihab
Foutse Khomh
In this work, we evaluate three popular fairness preprocessing algorithms and investigate the potential for combining all algorithms into a … (voir plus)more robust preprocessing ensemble. We report on lessons learned that can help practitioners better select fairness algorithms for their models.
2023-04-01
Computer (publié)
doi.org
arxiv.org
Deep learning-enabled anomaly detection for IoT systems
Adel Abusitta 0001
Adel Abusitta
Glaucio H.S. Carvalho
Omar Abdel Wahab
Talal Halabi
Benjamin Fung
Saja Al-Mamoori
2023-04-01
Internet of Things (publié)
doi.org
Facing AI extinction
David Scott Krueger
2023-04-01
New Scientist (publié)
doi.org
Genesis, modelling and methodological remedies to autism heterogeneity
Juliette Rabot
Eya‐mist Rødgaard
Ridha Joober
Guillaume Dumas
Danilo Bzdok
Boris C Bernhardt
Sébastien Jacquemont
Laurent Mottron
2023-04-01
Neuroscience and Biobehavioral Reviews (publié)
doi.org
Home alone: A population neuroscience investigation of brain morphology substrates
M. Noonan
Chris Zajner
Danilo Bzdok
2023-04-01
NeuroImage (publié)
doi.org
Investigating the neural correlates of affective mentalizing and their association with general intelligence in patients with schizophrenia
Wladimir Tantchik
M. J. Green
Yann Quidé
Susanne Erk
Sebastian Mohnke
Carolin Wackerhagen
Nina Romanczuk-seiferth
Heike Tost
Kristina Schwarz
Carolin Moessnang
Danilo Bzdok
Andreas Meyer-Lindenberg
Andreas Heinz
Henrik Walter
2023-04-01
Schizophrenia Research (publié)
doi.org
Machine-learning-based arc selection for constrained shortest path problems in column generation
Mouad Morabit
Guy Desaulniers
Andrea Lodi
Column generation is an iterative method used to solve a variety of optimization problems. It decomposes the problem into two parts: a maste… (voir plus)r problem and one or more pricing problems (PP). The total computing time taken by the method is divided between these two parts. In routing or scheduling applications, the problems are mostly defined on a network, and the PP is usually an NP-hard shortest path problem with resource constraints. In this work, we propose a new heuristic pricing algorithm based on machine learning. By taking advantage of the data collected during previous executions, the objective is to reduce the size of the network and accelerate the PP, keeping only the arcs that have a high chance to be part of the linear relaxation solution. The method has been applied to two specific problems: the vehicle and crew scheduling problem in public transit and the vehicle routing problem with time windows. Reductions in computational time of up to 40% can be obtained.
2023-04-01
INFORMS Journal on Optimization (publié)
doi.org
arxiv.org