FLFélix Camirand Lemyre

Félix Camirand Lemyre

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Félix Camirand Lemyre has a total of 82 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Operationalizing transparency regarding the use of health data for research: a cross-sectional survey of citizens’ perspectives

    by Annabelle Cumyn, Jean-Frédéric Ménard, Emmanuel Bilodeau, Roxanne Dault, Jean-Philippe Morissette, Marie-Pier Domingue, Félix Camirand Lemyre, Denis Boutin, Jean-François Ethier

    Background: Secondary use of health data can accelerate discovery and enable personalized care, yet social acceptability relies on autonomy, transparency, and trust, especially in the context of security breaches and concerns about misuse. In Quebec, as in many Western settings, there is no formal mechanism to inform individuals about such use, underscoring the need for practical solutions. This quantitative cross-sectional study assessed whether transparency mechanisms increase trust in the use of health data for research; examined their perceived priority and acceptability; and explored effects across population subgroups. Methods: We conducted a cross-sectional telephone survey among 366 adults in Quebec (Canada) selected via stratified random sampling. The sample included a random selection of adult household members based on age. Descriptive statistics, chi-square tests, and exploratory multivariate logistic regressions were performed. Results: More than half of respondents were unaware that their health data may be used for research. Trust was higher for public-sector than private-sector researchers. Overall, 76% indicated that access to information would increase their trust, particularly younger adults (OR = 1.98, 95% CI: 1.10–3.57). Over 90% judged implementing transparency mechanisms within five years as important, with stronger support among those aware of current data use (OR = 4.63, 95% CI: 1.33–16.20). A secure website was considered acceptable by 75% of respondents as a transparency mechanism. Individuals with high trust in public-sector researchers (OR: 2.06, 95% CI: 1.08–3.94) and younger adults (OR: 2.05, 95% CI: 1.13–3.71) were more likely to consider it important. Exploratory analyses suggested that awareness of current data use for research and pre-existing trust predict increased trust in researchers from various public and private organizations if a secure website is implemented. Conclusion: Providing information through a transparency mechanism, particularly via a secure website, appears acceptable, valued, and potentially effective for increasing trust in the use of health data for research. Our findings support the feasibility of a web-based transparency portal and suggest tailoring communication to demographic, trust and awareness profiles.

  • Open Access

    VALORIS : repenser les recettes pour l’analyse de données de santé distribuées

    by Marie-Pier Domingue, Félix Camirand Lemyre

    En combinant les multiples données de santé générées par une même personne, il devient possible de mieux comprendre les trajectoires du vieillissement. Pourtant, des contraintes légales et éthiques empêchent le partage de ces données individuelles entre organisations. Comme les méthodes usuelles pour analyser ces données ont été bâties pour le cadre dans lequel les données sont centralisées, les recettes existantes pour ces analyses ont dû être entièrement repensées afin de développer des méthodes dites distribuées. C'est ainsi qu'est né VALORIS, un algorithme distribué pour le modèle de régression logistique. VALORIS ouvre la voie à de nouvelles possibilités pour la recherche concernant les trajectoires du vieillissement. Marie-Pier Domingue est doctorante en mathématiques et en biostatistiques au sein du Groupe de recherche interdisciplinaire en informatique de la santé (GRIIS), en cotutelle entre l’Université de Sherbrooke et l’Université Paris Cité. Félix Camirand Lemyre, chercheur au Centre de recherche sur le vieillissement et professeur agrégé en mathématiques à l’Université de Sherbrooke. Il est codirecteur scientifique du GRIIS et directeur du Centre de consultation statistique de l’Université de Sherbrooke. Il est titulaire de la chaire de recherche en méthodologies statistiques appliquées aux enjeux de santé et phénomènes sociaux.

  • Open Access

    Nonparametric Density Estimation of a Long-Term Trend from Repeated Semicontinuous Data

    by Félix Camirand Lemyre, Raymond J. Carroll, Aurore Delaigle

    We consider nonparametric estimation of the density of the long-term trend of a semicontinuous variable observed repeatedly over time. These variables arise when measuring the intensity of an intermittent phenomenon, such as the intake of an episodically consumed nutrient or the concentration of an intermittent toxic substance: when the phenomenon is absent, the measurement is equal to zero; otherwise, it is positive. Semicontinuous data are usually represented by a two-part model describing the zeros and the nonzeros separately, often under parametric assumptions. Recently, Camirand Lemyre, Carroll, and Delaigle showed that it is possible to relax the distributional assumptions on the part that models the nonzeros, but like other existing work, they used a parametric model for the conditional probability H of observing a nonzero value. We develop a nonparametric estimator of H and of the density of the long-term trend. We illustrate our method on simulated examples and apply it to estimate the density of long-term fruit intake, using data from the Eating at America’s Table Study. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

  • Open Access

    Modeling Temporal Summation and Conditioned Pain Modulation in Individuals with and without Chronic Pain.

    by Matthieu Vincenot, Simon Lévesque, Louis Gendron, Félix Camirand Lemyre, Serge Marchand, Guillaume Léonard

    Although significant progress has been made in recent years in the field of pain modulation, information regarding patients' pain modulation profiles remains largely research-bound and is not yet easily accessible in clinical settings. The aim of this study was to develop and validate a model for estimating pain modulation profiles - including temporal summation of pain (TSP) and conditioned pain modulation (CPM) - using easily accessible measures. This study included 347 pain-free individuals and 108 with chronic pain. TSP was induced via tonic heat pain stimulations, while CPM was evaluated using pressure pain thresholds (CPM-PPT) and tonic heat pain stimulations (CPM-HPS) as test stimuli, with the cold pressor test as the conditioning stimulus. Independent variables included demographic, psychological, and physiological measures. A LASSO regression with cross-validation was used to identify key independent. For TSP, the model explained 40% of the variance, incorporating factors such as monoamines and blood pressure. For CPM, the CPM-HPS model performed best, accounting for 35% of the variance, with blood pressure, sex and pain catastrophizing identified as important predictors. Overall, these results indicate that TSP and CPM can be partially estimated using readily accessible measures, but high prediction error currently limits their clinical applicability.


  • Open Access

    VALORIS: A privacy-aware logistic regression method for vertically partitioned data within a novel privacy risk assessment framework

    by Félix Camirand Lemyre, Marie-Pier Domingue, Jean-Philippe Morissette, Anita Burgun, Jean-François Ethier

    Life sciences research increasingly relies on variables held by different entities, such as clinical, laboratory, environmental, and genomic data. Due to legal, ethical, and social acceptability constraints, these data often cannot be shared across organizations holding them. As a result, they cannot be pooled, and analyses must be conducted within the framework of vertically partitioned data. Supporting such analyses requires methods that protect privacy. However, the mere fact that line-level data are not exchanged should not be mistaken for true privacy protection. We introduce VALORIS (Vertically partitioned Analytics under the LOgistic Regression model for Inference in Statistics), a novel method that enables statistical inference under a logistic regression model without disclosing any individual-level data—including the outcome variable. VALORIS is a practical, communication-efficient algorithm that requires no third-party coordinator. Most importantly, it includes a novel framework for evaluating privacy, allowing users to distinguish among different levels of privacy preservation.

  • Open Access

    Revisiting VERTIGO and VERTIGO-CI: Identifying confidentiality breaches and introducing a statistically sound, efficient alternative

    by Marie-Pier Domingue, Jean-François Ethier, Jean-Philippe Morissette, Simon Lévesque, Anita Burgun, Félix Camirand Lemyre

    Objectives: This study aims to guide future use of the VERTIGO and VERTIGO-CI algorithms by (1) Highlighting confidentiality issues; (2) minimizing the number of communication rounds required; (3) offering an alternative algorithm that excludes the intercept from the penalty. Materials and Methods: Quantities exchanged in the original algorithms were inspected and linear algebra was used to uncover reverse-engineering processes that the coordinating center can perform to re-create raw data. The objective function of the optimization problem was investigated to propose an alternative where one communication is needed to run the algorithm while being able to exclude the intercept from the penalty. Results: It was shown that the coordinating center can retrieve all individual data using matrix operations when VERTIGO-CI is run. Individual data could also be retrieved in the case where parameter estimates associated with binary covariates are shared. The VERTIGO algorithm was adapted with reduced communications, and an algorithm was proposed to exclude the intercept from the penalty. Discussion: While the use of VERTIGO-CI or VERTIGO with binary covariates does not imply directly sharing raw data, this works highlights the risks of reverse-engineering enabling the coordination node to effectively obtain the raw dataset of a data node. While inspecting quantities exchanged allowed to reduce the number of communications rounds, investigating the modeling choice, which translates as including or not the intercept in the penalty, was essential for interpretation matters. Conclusion: This work highlights confidentiality breaches that may arise in distributed analytics methods, but also proposes an improved algorithm with reduced operational costs and enriching model interpretablity.

  • Open Access

    Exploring the Spectrum of Temporal Summation and Conditioned Pain Modulation Responses in Pain‐Free Individuals Using a Tonic Heat Pain and Cold Pressor Test Paradigms

    by M. Vincenot, G. Léonard, C. Cloutier‐Langevin, M. Bordeleau, L. Gendron, F. Camirand Lemyre, S. Marchand

    Background: Temporal summation (TS) and conditioned pain modulation (CPM) are impaired in many chronic pain individuals, although typical TS and CPM responses remain unclear. Providing a quantile distribution of TS and CPM values could help determine whether TS and CPM responses fall outside the usual ranges, thereby informing pain management strategies. The objective of this study was to provide a response distribution for these mechanisms in a large, pain‐free cohort.Methods: Pain‐free participants (n = 347) were evaluated. TS was interpreted as the change in pain perception scores during a constant heat pain stimulus. CPM was assessed by the difference in pain perception scores induced by a tonic heat pain stimulus and a mechanical pain threshold, before and after a cold pressor test (CPT). Quantile regression was performed to identify the 5th to 95th percentiles, with a specific focus on the 75th, 90th and 95th percentiles and their confidence intervals, while considering age and sex.Results: Analyses revealed a particularly large range of TS and CPM responses, spanning from strong inhibition in CPM and strong summation in TS to moderate or even null effects. Notably, some responses deviated from the expected patterns, with certain individuals exhibiting hyperalgesic CPM effects and others displaying hypoalgesic TS responses. The highest percentiles of the TS distribution were influenced by age and gender, while CPM scores remained unchanged.Conclusion: The results of this study provide an overview of the possible spectra of responses for TS and CPM, potentially paving the way for their incorporation into pain management strategies.Significance Statement: This study examines the spectrum and provides a distribution for TS and CPM values to better identify what may constitute a deficit in these mechanisms and enhance their clinical interpretation.Trial Registration: clinicaltrial.gov identifier: NCT03376867


  • Open Access

    Correction: Distributed Statistical Analyses: A Scoping Review and Examples of Operational Frameworks Adapted to Health Analytics

    by Félix Camirand Lemyre, Simon Lévesque, Marie-Pier Domingue, Klaus Herrmann, Jean-François Ethier

    Background: Data from multiple organizations are crucial for advancing learning health systems. However, ethical, legal, and social concerns may restrict the use of standard statistical methods that rely on pooling data. Although distributed algorithms offer alternatives, they may not always be suitable for health frameworks. Objective: This study aims to support researchers and data custodians in three ways: (1) providing a concise overview of the literature on statistical inference methods for horizontally partitioned data, (2) describing the methods applicable to generalized linear models (GLMs) and assessing their underlying distributional assumptions, and (3) adapting existing methods to make them fully usable in health settings. Methods: A scoping review methodology was used for the literature mapping, from which methods presenting a methodological framework for GLM analyses with horizontally partitioned data were identified and assessed from the perspective of applicability in health settings. Statistical theory was used to adapt methods and derive the properties of the resulting estimators. Results: From the review, 41 articles were selected and 6 approaches were extracted to conduct standard GLM-based statistical analysis. However, these approaches assumed evenly and identically distributed data across nodes. Consequently, statistical procedures were derived to accommodate uneven node sample sizes and heterogeneous data distributions across nodes. Workflows and detailed algorithms were developed to highlight information sharing requirements and operational complexity. Conclusions: This study contributes to the field of health analytics by providing an overview of the methods that can be used with horizontally partitioned data by adapting these methods to the context of heterogeneous health data and clarifying the workflows and quantities exchanged by the methods discussed. Further analysis of the confidentiality preserved by these methods is needed to fully understand the risk associated with the sharing of summary statistics.

  • Open Access

    Distributed Statistical Analyses: A Scoping Review and Examples of Operational Frameworks Adapted to Health Analytics

    by Félix Camirand Lemyre, Simon Lévesque, Marie-Pier Domingue, Klaus Herrmann, Jean-François Ethier

    Background: Data from multiple organizations are crucial for advancing learning health systems. However, ethical, legal, and social concerns may restrict the use of standard statistical methods that rely on pooling data. Although distributed algorithms offer alternatives, they may not always be suitable for health frameworks. Objective: This study aims to support researchers and data custodians in three ways: (1) providing a concise overview of the literature on statistical inference methods for horizontally partitioned data, (2) describing the methods applicable to generalized linear models (GLMs) and assessing their underlying distributional assumptions, and (3) adapting existing methods to make them fully usable in health settings. Methods: A scoping review methodology was used for the literature mapping, from which methods presenting a methodological framework for GLM analyses with horizontally partitioned data were identified and assessed from the perspective of applicability in health settings. Statistical theory was used to adapt methods and derive the properties of the resulting estimators. Results: From the review, 41 articles were selected and 6 approaches were extracted to conduct standard GLM-based statistical analysis. However, these approaches assumed evenly and identically distributed data across nodes. Consequently, statistical procedures were derived to accommodate uneven node sample sizes and heterogeneous data distributions across nodes. Workflows and detailed algorithms were developed to highlight information sharing requirements and operational complexity. Conclusions: This study contributes to the field of health analytics by providing an overview of the methods that can be used with horizontally partitioned data by adapting these methods to the context of heterogeneous health data and clarifying the workflows and quantities exchanged by the methods discussed. Further analysis of the confidentiality preserved by these methods is needed to fully understand the risk associated with the sharing of summary statistics.


  • Implementing pathogen reduction technology while discontinuing blood donor deferral criteria for sexual risk behaviors: A simulation study

    by Marie‐Pier Domingue, Sheila F. O'Brien, Yves Grégoire, Marion C. Lanteri, Susan L. Stramer, Félix Camirand Lemyre, Antoine Lewin

    Background: Combining pathogen reduction technology (PRT) with blood screening may alleviate concerns over the risk of transfusion‐transmitted infections (TTI) and support changes in blood donor selection to potentially increase blood availability. This study aimed to estimate the residual risk of human immunodeficiency virus (HIV), hepatitis B virus (HBV), and hepatitis C virus (HCV) transfusion‐transmission in Canada after implementing PRT, while eliminating deferrals for sexual risk behaviors.Study Design and Methods: A probabilistic approach that combined Bayesian networks with Monte Carlo simulations was used to estimate the risk of transfusing HIV‐, HBV‐, or HCV‐contaminated blood components. Different scenarios were considered to compare the current residual risk after PRT implementation, with and without donor deferral criteria for sexual risk behaviors. Donor profiles and blood component outcomes were simulated based on a literature review including the prevalence and incidence of HIV, HBV, and HCV in the Canadian blood donor population; the use of current blood screening assays; and HIV, HBV, and HCV blood donor viral loads.Results: In the universal PRT scenario (i.e., with PRT/without deferral criteria), the estimated risks of HIV, HBV, and HCV transmission were significantly lower than those in the currently observed scenario (i.e., without PRT/with deferral criteria).Conclusions: This risk model suggests that PRT for platelets and plasma (and eventually for RBCs when available) significantly reduces the residual risks of HIV, HBV and HCV transfusion‐transmission and could enable the removal of blood donor deferral criteria for sexual risk behaviors.