Team
Image Processing and Pattern RecognitionCampus
ParisContact
michael.atlan@epita.fr
Short Bio
Michael Atlan is a researcher at the French National Center for Scientific Research in physics, instrumentation, and computational imaging. He works at the Langevin Institute, the Quinze-Vingts National Eye hospital, and the Adolphe de Rothschild Foundation Hospital in Paris, France, and is also an Associate Researcher with the Image Processing and Pattern Recognition group. He pioneered the development and global deployment of Doppler imaging for ophthalmology via high-bitrate digital holography, combining advanced engineering with open-source, high-performance software. He leads the establishment of quantitative metrics and standardized biomarkers to enable large-scale clinical trials in ophthalmology and cardiovascular medicine. He is the project manager of the versatile real-time digital hologram rendering software Holovibes and the computational image rendering prototyping platform Holowaves. He is the president of the digital holography foundation.
Research
Ultrafast retinal Doppler holography (https://arxiv.org/abs/1804.10066) is becoming a reference, noninvasive functional exam of the retinal microcirculation. Performed in a few seconds on an undilated eye with diffuse near-infrared illumination(https://arxiv.org/abs/2212.13347), this exam reconstructs arterial and venous flow-velocity waves that reflect variations in ocular perfusion pressure (https://arxiv.org/abs/2409.17180).
This approach provides access to dynamical metrics (blood-flow volume, vessel-wall motion), mechanical metrics (resistance, compliance, elasticity), and rheological metrics (viscosity, wall shear stress) unprecedented in clinical practice. An open-source software pipeline enables high-throughput, real-time acquisition (https://arxiv.org/abs/2508.03911, https://github.com/DigitalHolography/Holovibes), image rendering with aberration compensation (https://doi.org/10.1364/BOE.528568), automated extraction of quantitative metrics and biomarkers (https://github.com/DigitalHolography/HoloDoppler), and generation of a standardized clinical report (https://github.com/DigitalHolography/EyeFlow/) - the foundation for ongoing multicenter clinical studies across the United States, Europe, and China.
These rigorous, standardized metrics offer powerful tools for triage, personalized treatment adjustment, and longitudinal follow-up of patients and large-scale cohorts. They directly inform studies of major blinding diseases such as glaucoma, diabetic retinopathy, retinal vein occlusion, and ocular ischemia, while extending their relevance to systemic conditions such as cardiovascular disease.
Tracking the hemodynamic, mechanical, and rheological signatures of the retinal vascular bed provides an objective language of microvascular health and vascular risk. Our ambition is to identify and validate biomarkers that enable intervention at the earliest stages of microvascular impairment, when physiology remains reversible, to prevent the onset of structural lesions rather than observe them a posteriori.