Digital Retinal Screening Reveals State of Health

Digital techniques and big data can provide an accurate picture of a person's general medical condition by imaging the retina, claims a new study.

Under development at the Medical University of Vienna (MedUni; Austria), the automated digital retinal screening technique uses optical coherence tomography (OCT) technology to perform 40,000 scans of the retina within 1.2 seconds, with a total volume of 65 million voxels. The OCT data are then analyzed using automated algorithms generated on the basis of artificial intelligence (AI), revealing a range of information on the subject, such as age, gender, smoking habits, blood pressure, and diseases or susceptibility to diseases such as diabetes.


According to the researchers, the innovative approach represents an important step towards better comprehensive care of the rapidly increasing number of diabetics, and that it might be possible in the future to also use OCT to detect diseases of some internal organs, such as the kidneys, or age-related problems and neurological diseases from a retina scan. Both the device and the AI algorithms were presented at the Advanced Retinal Therapy (ART) specialist conference, held during December 2017 in Vienna (Austria).

“As a window into the condition of a person's brain and vascular system, the retina offers huge insights into their life science data; not only about existing or potential diseases, but also about lifestyle,” said Ursula Schmidt-Erfurth, MD, head of the MedUni department of ophthalmology and optometry. “A large number of digital techniques are already being used in the management of eye conditions, always with the aim of improving standards of ophthalmological care. Digital retinal scanning is an additional revolutionary step in this direction.”

“Diabetes and high blood-pressure are very common conditions and, in many patients, result in long-term retinal damage. It has hitherto not been possible for us, as internists, to look into the eye, because we lacked diagnostic experience and did not have the necessary equipment in our department,” said Florian Kiefer, MD, of the MedUni department of medicine. “By incorporating these new technologies into clinical care, we will be able to obtain a much more accurate picture of our patients' general health, so that we will not only be able to offer them individual information and advice, but also customized treatments.”

OCT is based on low-coherence interferometry, typically employing near-infrared (NIR) light. The use of relatively long wavelength light allows it to penetrate into the scattering medium. Depending on the properties of the light source, OCT can achieve sub-micrometer resolution. OCT, being an echo imaging method, is similar to ultrasound imaging, but can image only 1-2 mm below the surface of biological tissue, as at greater depths the proportion of light that escapes without scattering is too small to be detected.

Related Links:
Medical University of Vienna