Artificial Intelligence and Advancing Eyecare

For a long time humanity has looked towards a future of technology. Although we have only scratched the surface, artificial intelligence has taken over most aspects of our daily lives. Artificial Intelligence (AI) has drastically changed the landscape of medical imaging and diagnoses; with the advancement of AI and technology, ocular diseases, many of which go undiagnosed, can be detected and treated early. The more common diseases such as Glaucoma, Age-related Macular Degeneration (AMD), and Diabetic Retinopathy can be detected with comprehensive eye exams and if performed regularly can provide early detection of other diseases such as diabetes, arthritis, and cancer.

Cristina Gonzalez Gonzalo, PhD Candidate at the A-eye Research and Diagnostic Image Analysis Group of Radboudumc, has worked to make AI systems more applicable to clinical practices. Typically eye scans are able to detect deformities in the retina, including AMD and diabetic retinopathy, however AI systems are limited in the sense that their analysis will only obtain just enough information to make the diagnosis, deeming it oblivious to other abnormalities that the system failed to pick up. The development of her “iterative process” forces the AI system to more transparently scan images; avoiding areas that have already been detected and show parts of the eye where new abnormalities may exist. To read more about Gonzalo’s innovation in her attempt to make AI-systems more relevant to clinical practices refer to the article on:

https://www.sciencedaily.com/releases/2020/06/200624120434.htm

Furthermore, software based on artificial intelligence has also been developed by researchers at the Eye Clinic of the University Hospital Bonn, Stanford University and the University of Utah that allows for precise analysis of Geographic Atrophy (GA). Geographic Atrophy is an advanced form of age-related macular degeneration (AMD) in which there is a loss of retinal cells (photoreceptors, epithelial cells, and vessels) resulting in partial or complete loss of vision. Vision with and without age-related macular degeneration is shown in Figure 1. GA can be detected by an ophthalmologist during a dilated exam, and further retinal imaging will show what is represented in Figure 2.

Figure 1. https://onlinelibrary.wiley.com/doi/full/10.1111/opo.12419

Figure2.https://www.reviewofophthalmology.com/article/chasing-geographic-atrophy-treatments

The approach developed by researchers allows for intricate analysis of lesions as well as three-dimensional visualization of the structure of the retina. Apart from imagining the lesions, accurate analysis of the light-sensitive cells surrounding them will serve as a predictor for future progression of the disease. The use of this AI system in clinical practices will essentially help ocular therapy of surrounding intact light sensitive cells and possibly slow down the progression of the disease that currently has no effective treatment. To learn more about the development of this AI system visit

https://www.sciencedaily.com/releases/2020/08/200813123552.htm

Just to leave you with a real idea of how advanced research in AI is in the field of Eye Care, an international team led by scientists at the Hong Kong University of Science and Technology (HKUST) have developed the world’s first 3D artificial eye with capabilities that move beyond the current bionic eyes. What distinguishes it from bionic eyes is a 3D artificial retina made of nanowire light sensors that replicate photoreceptors in the human eyes, and also have the potential to receive more light signals than the human retina. In partnership with the University of California, Berkeley researchers are looking to work on the biocompatibility which will allow for prosthetic application and use in clinical practices. You can learn more about this project at

https://www.sciencedaily.com/releases/2020/06/200610102726.htm

---

References

Dunaief, J., & Institute, S. E. (2020, July 30). What is Geographic Atrophy? Retrieved October

17, 2020, from https://www.brightfocus.org/macular/article/what-geographic-atrophy

Hong Kong University of Science and Technology. (2020, June 10). World's first spherical

artificial eye has 3D retina. ScienceDaily. Retrieved October 17, 2020 from

Radboud University Medical Center. (2020, June 25). Towards an AI diagnosis like the

doctor's: How can we make 'lazy' artificial intelligence more transparent and relevant to

the clinic?. ScienceDaily. Retrieved October 17, 2020 from

www.sciencedaily.com/releases/2020/06/200624120434.htm

University of Bonn. (2020, August 13). Artificial intelligence recognizes deteriorating

photoreceptors. ScienceDaily. Retrieved October 17, 2020 from

www.sciencedaily.com/releases/2020/08/200813123552.htm