Simon de Montigny

Simon de Montigny
Université de Montréal | UdeM · School of Public Health

PhD

About

17
Publications
706
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
31
Citations
Additional affiliations
January 2015 - present
Université de Montréal
Position
  • PostDoc Position
August 2007 - August 2014
Polytechnique Montréal
Position
  • PhD Student
May 2006 - present
Polytechnique Montréal
Position
  • Lecturer

Publications

Publications (17)
Preprint
BACKGROUND: Mathematical models based on the physiology when programmed as a software can be used to teach cardiorespiratory physiology and to forecast the effect of various ventilatory support strategies. We developed a cardiorespiratory simulator for children called “SimulResp”. The purpose of this study was to evaluate the quality of SimulResp....
Article
Full-text available
Background: The rise of big data and related predictive modelling based on machine learning algorithms over the last two decades have provided new opportunities for disease surveillance and public health preparedness. Big data come with the promise of faster generation of and access to more precise information, potentially facilitating predictive...
Chapter
Neural networks have been investigated as models for survival data using a training criterion similar to that of the Cox proportional hazards model, a criterion not designed for clinical prediction. In this paper, we develop a new survival learning algorithm where a neural network ensemble minimizes the integrated Brier score. We compare the result...
Article
Full-text available
Promising multi-dose HIV vaccine regimens are being tested in trials in South Africa. We estimated the potential epidemiological and economic impact of HIV vaccine campaigns compared to continuous vaccination, assuming that vaccine efficacy is transient and dependent on immune response. We used a dynamic economic mathematical model of HIV transmiss...
Article
Full-text available
Background: The epidemiological tipping point ratio (TPR) has been suggested as a useful indicator to monitor the scale-up of antiretroviral (ART) programmes and determine when scale-up is sufficient to control the epidemic. TPR has been defined as the ratio of yearly number of new HIV infections to the yearly number of new ART initiations or to th...
Poster
Link to abstract and poster: http://www.croiconference.org/sessions/designing-hiv-vaccine-delivery-strategies-south-africa-policy-analysis
Poster
Link to abstract and poster: http://www.croiconference.org/sessions/assessing-utility-tipping-point-ratio-monitoring-art-program-success
Article
In this paper, we obtain simplified as well as original closed-form expressions for integrals involving the gamma function. These explicit results are found neither in the literature nor using symbolic computation software. Subsequent results follow, giving rise to explicit expressions for integrals involving the error function, with application in...
Article
Error backpropagation in networks of spiking neurons (SpikeProp) shows promise for the supervised learning of temporal patterns. However, its widespread use is hindered by its computational load and occasional convergence failures. In this letter, we show that the neuronal firing time equation at the core of SpikeProp can be solved analytically usi...
Article
In this work, we propose a new approximation method to perform error backpropagation in a quantron network while avoiding the silent neuron problem that usually affects networks of realistic neurons. In our experiments, we train quantron networks to solve the XOR problem and other nonlinear classification problems. We achieve this while using less...
Article
The quantron is a hybrid neuron model related to perceptrons and spiking neurons. The activation of the quantron is determined by the maximum of a sum of input signals, which is difficult to use in classical learning algorithms. Thus, training the quantron to solve classification problems requires heuristic methods such as direct search. In this pa...
Conference Paper
The quantron is a new artificial neuron model, able to solve nonlinear classification problems, for which an efficient learning algorithm has yet to be developed. Using surrogate potentials, constraints on some parameters and an infinite number of potentials, we obtain analytical expressions involving ceiling functions for the activation function o...

Network

Cited By

Projects

Projects (2)
Project
Decision support systems in clinical care and public health are based on models and simulations (M&S) that predict the outcomes of envisioned interventions based on explicit biological or social mechanisms that explain the occurrence of events or the evolution of processes of relevance. There is great interest in embedding machine learning into M&S components of decision support systems to handle big health data with the main objectives of (1) improving the accuracy of outcome predictions by fully utilizing the context in which interventions are embedded, (2) reducing the time and complexity currently required to develop core M&S components. The difficulty of merging M&S and machine learning is that of interpretability, a quality which is manifest in M&S as a result of the deliberations of modelers, engineers, and health scientists, but not so in machine learning where there is a disconnect between the capacity of computers and algorithms to make predictions and that of machine learning users to understand the basis on which these predictions are made. This proposal targets the urgent need to integrate big health data into a decision support system in a systematic manner. Our objective is to develop a machine learning modeling and simulation (ML-M&S) framework for that purpose, with application in cardiorespiratory critical care.