On April 8th, 2025, Dr. Hugo Le Boité successfully defended his PhD thesis titled “Deep Learning tools for evaluation of diabetic retinopathy and maculopathy”. This marked the culmination of a three and a half-year journey during which Dr. Le Boité developed and applied tools for the segmentation of retinal non-perfusion in OCT-Angiography images in diabetic retinopathy, tools for the segmentation of pathological fluids in OCT images, as well as innovative methodological approaches to alleviate the manual annotation burden when constructing large database of medical images in the purpose of training efficient deep learning models.
The defense was held in the presence of a jury comprising the following members:
- Mr. Raphael Sznitman: professor and research director at the ARTORG center for biomedical engineering and research, Bern Switzerland.
- Ms. Alexandre Miere: assistant professor of ophthalmology in the ophthalmology department of the Creteil Hospital (Centre Hospitalier Intercommunal de Créteil, France)
- Mr. Fabrice Meriaudeau: professor, and research director at the Medical Imaging group, Université de Bourgogne. President.
- Mr. Xavier Guillonneau: research director at the Institut de la Vision, Sorbonne Université, Paris, France.
- Ms. Catherine Creuzot Garcher: professor, ophthalmology department of the Dijon University Hospital, France.
- Mr. Gwenolé Quellec: research director, INSERM, LaTIM, UMR-1101.
- Ms. Aude Couturier: professor, ophthalmology department of the Lariboisière hospital, Paris, France.
This research provides potential deep learning tools that could be applied in clinical practice for automatic segmentation of specific biomarkers, but also provides interesting methodological techniques that will be applied in future projects in order to optimize manual annotation in the production of medical image segmentations.
Appearing in the top picture, from left to right: Fabrice Meriaudeau, Hugo Le Boité, Xavier Guillonneau, Gwenolé Quellec, Aude Couturier, Alexandra Miere. Catherine Creuzot Garcher and Raphael Sznitman participated remotely.
