top of page
LiGHT Safe Area.png
Asset 25.png

Publications

LiGHT designs, deploys, and rigorously evaluates AI systems through large-scale international clinical trials, particularly embedded in low-resource and high-variability environments. As an academic lab, we emphasize methodological rigor, reproducibility, and evidence generation.

Asset 25.png

Publications

Publications

IEEE

2025

TimEHR: Image-based Time Series Generation for Electronic Health Records

Hojjat Karami, Mary-Anne Hartley, David Atienza, Anisoara Ionescu4

Society of Nuclear Medicine

2025

Single-Center Evaluation of Meditron - a Large Language Model for Nuclear Medicine Clinical Real-World Scenarios: A Performance Analysis

Mario Jreige, Giorgia Carra, Marie Nicod lalonde, Niklaus Schaefer, Mary-anne Hartley, Noémie Boillat blanco, Francois Bastardot, Alexandre Sallinen, John Prior, Patric Hagmann and Jean louis Raisaro

SCIENCE DIRECT

2025

InfoKids+: A Validation Study of a Pediatric Acuity Risk Stratification Algorithm

Carl A. Starvaggi, Sophie Affentranger, Noelie Lengeler, Johan N. Siebert, Annick Galetto-Lacour, Rainer Tan, Manon Jaboyedoff, Claudia E. Kuehni, Mary-Anne Hartley, Kristina Keitel

Logo Long Black.png

EPFL, Lausanne, Switzerland Harvard, Ariadne Labs, Boston, USA

Logo Long Black.png

Home

Who we are

Contact

EPFL, Lausanne, Switzerland Harvard, Ariadne Labs, Boston, USA.

mary-anne.hartley@ epfl.ch

bottom of page