Periodic Reporting for period 1 - GSAORI (Guide Stars Adaptive Optics for Retinal Imaging)
Berichtszeitraum: 2016-04-01 bis 2018-03-31
Although adaptive optics has enabled outstanding progresses in ophthalmology, it has its limitations. Pupil motion, tear film thickness changes during the examination, the structure of the diseased retina and permanent eye motion can limit the performance of the system. Therefore satisfactory image quality can only be achieved in healthy volunteers or carefully selected patients, which currently prevents a widespread use of this technology. Moreover, only small part of the retina can be imaged with classical adaptive optics systems, due to the variation of the imperfections within the eye. The ability to image larger patches of the retina would greatly improve the clinical interest of this technique.
The overall objective of the project is to overcome these difficulties by developing more robust and more efficient adaptive optics systems. This will enable reasonable image quality in the majority of patients, on a relatively large field of view and in a short time. The emergence of such tool would be a significant milestone towards making this technique a routine for eye examination and, thus, would greatly improve diagnosis capabilities and reduce serious consequences of diseases that are diagnosed too late.
The first part of the project was focused on developing end-to-end modelling techniques. Indeed, the simulation of adaptive optics systems is important in order to predict their performance, and thus optimize them. The developed modelling techniques were really helpful in the second phase of the project, which consisted in the conception of a new instrument for large field, high resolution retinal imaging. Such an instrument is robust, fast and is providing images of patches of the retina that are four times larger than in classical adaptive optics instruments. The third part of the project was experimental. The new instrument was gradually built, each iteration increasing its performance. In-vivo measurements of the eyes of healthy volunteers and patients were carried out to validate the capabilities of the system.
The attached figure gives an example of retinal images obtained on one healthy volunteer. A patch of 1.5 mm x 1.5 mm located 2.5° away from the fovea (center of vision) was imaged here. The figure presents images of different retinal layers (photoreceptor layer, anterior blood vessel layer and nerve fiber layer). Individual cone photoreceptors can be identified, as well as small capillaries and nerve bundles. The small dimension of these details is well demonstrating the high resolution of our instrument.
In conclusion, the developed instrument provides high resolution images of the retina on a relatively large field of view, in a short time (less than one minute). The complexity of the system is moderate and its operation simple.
During this project, the researcher published 4 papers in the journal Biomedical Optics Express (2 as first authors). The scientific results were also presented during talks in 3 international conferences. Finally, the researcher gave 2 seminars about the research related to this project.