2023-… deputy director of the BIOSTAT research team, Inserm BPH, Bordeaux, France.
2019-… Director of Research, Inserm, Bordeaux, France.
2008-2019 Researcher, Inserm, Bordeaux, France.
2008 Research Assistant, Biostatistics and Nutrition departments, Inserm, Bordeaux, France.
2008 Assistant Professor, ISPED, Université Bordeaux Segalen, Bordeaux, France.
2007 Postdoctoral fellow, Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor, USA.
2003-2006 Assistant professor, ISPED, Université Bordeaux Segalen, Bordeaux, France.
2014 HDR (Habilitation to conduct research), Univ. Bordeaux, France.
2006 PhD in Biostatistics, Univ. Bordeaux Segalen, France.
2003 Master degree in Epidemiology and Public Health, ISPED, Univ. Bordeaux Segalen, France.
2002 Master degree in Biostatistics, National School for Statistics and Information Analysis, ENSAI, Rennes, France.
Development of statistical methods to describe, explain and predict chronic disease progressions.
I specialized over the years in latent class and latent process models for the joint analysis of correlated longitudinal markers and event time history with applications notably in cerebral aging and neurodegenerative diseases (Alzheimer’s Disease and related dementias, Multiple System Atrophy).
Beyond the development and application of dynamic statistical models, my group also develops individual dynamic predictions to quantify the risk of experiencing a clinical event based on the dynamic history of an individual, causal inference techniques for longitudinal data, and composite methodologies combining biostatistical modelling and machine learning methods.
All the developments, made available through open-source software (e.g., R packages lcmm, DynForest), are intended to address Public Health research questions through close collaborations with epidemiologists and clinicians and the analysis of large epidemiological cohort studies.
Joint models, Latent class mixed models, Latent variable models, (Generalized) linear mixed models, Longitudinal analysis of psychometric scales in Masters, summer schools, short courses
I serve as Associate Editor for Biometrics (since 2014) and I am a former Associate Editor of Biostatistics (2016-2022).
I belong to the leadership committee of the MELODEM initiative (Methods in Longitudinal Research in Dementia). http://melodem.org/
I am part of the STRATOS initiative (STRengthening Analytical Thinking for Observational Studies), topic group Measurement Error and Misclassification since 2022. https://stratos-initiative.org/
I am a member of the Representative Council of the International Biometric Society (IBS) and of the Council of the French Biometric Society (SFB).
See on WOS: https://www.webofscience.com/wos/author/record/1208723
or ORCID: https://orcid.org/0000-0002-9884-955X
For pre-prints, see Arxiv page: http://arxiv.org/a/proustlima_c_1
lcmm: Extended Mixed Models Using Latent Classes and Latent Processes. R package: https://cecileproust-lima.github.io/lcmm/
DynForest: Random Forest with Multivariate Longitudinal Predictors. R package https://github.com/anthonydevaux/DynForest
hdlandmark: landmark approach for dynamic prediction of time-to-events from large dimensional longitudinal biomarker history. R package; https://github.com/anthonydevaux/hdlandmark
marqLevAlg: A Parallelized Algorithm for Least-Squares Curve Fitting. R package; https://github.com/VivianePhilipps/marqLevAlgParallel
JLPM: Joint models based on Latent Processes. R package; https://github.com/VivianePhilipps/JLPM
CInNPL: Causal Inference in a Network of Latent Processes. R package; https://github.com/bachirtadde/CInLPN
NormPsy: Normalisation of Psychometric Tests. R package
See my github repository at https://github.com/CecileProust-Lima
2024 Analysis of multivariate longitudinal and survival data: from joint models to random forests – updated
2023 Analysis of multivariate longitudinal and survival data: from joint models to random forests
2023 Correctly accounting for misclassification when linking latent groups with external variables
2023 How to include time-varying exposures prone to measurement error in survival analyses?
2022 Joint Models based on Latent Classes and Latent Processes using lcmm R package
2021 Joint modelling of multivariate markers measured repeatedly over time and clinical endpoints
