RetinaLyze Receives Funding to Detect Diseases Such as Alzheimer’s Using AI

Source: RetinaLyze

RetinaLyze System A/S (Ltd.), a Denmark-based medical technology company that specializes in retinal imaging screening software, has received funding from The European Regional Development Fund (ERDF) and Central Denmark Region. The funding will be used to explore and develop novel AI-based solutions for automated disease and pathology screening/detection. RetinaLyze System’s existing automated screening solution and telemedicine module enables eye-screening professionals to perform efficient, accessible and fast eye-screenings for diabetic retinopathy (DR), AMD and glaucoma. Customers using the system have found several cases of serious and life-threatening diseases needing immediate treatment including tumors, edemas, arterial hypertension and benign intracranial hypertension apart from DR, AMD, and glaucoma.

Recent research indicates that there are correlations between common degenerative eye diseases, such as DR/AMD/glaucoma and Alzheimer’s. According to a study involving 3,877 randomly selecting patients by the University of Washington School of Medicine, patients with age-related macular degeneration, diabetic retinopathy or glaucoma have a 40% to 50% greater risk of Alzheimer’s disease than those without the eye conditions.

Inspired by these and similar developments in research, RetinaLyze System is actively working to expand the capabilities of it’s AI-software to detect more diseases, including diseases not directly related to the retina of the eye.

Cooperation with two universities and funding from ERDF

RetinaLyze System is participating in the project ‘Innovative use of Big Data: Deep Learning-based image analysis’ to further pursue this goal. The project is funded by The European Regional Development Fund (ERDF) and Central Denmark Region. The initiative helps companies with large amounts of data explore new ideas and develop new products using artificial intelligence (AI). The other project partners are two knowledge institutes: Visual Computing Lab from Alexandra Institute (owned by the Aarhus University Research Fund) and the Image analysis group at the Institute for Architecture and Media technology from Aalborg University as well as three innovative Danish tech companies.

As part of its ongoing work in preventing blindness and increasing the efficiency of eye specialists, RetinaLyze has built a significantly sized database of retinal images which can be used to train the new AI models. Apart from our own data, several public datasets will be included in the project to further increase the clinical accuracy and holistic approach of the AI-models.

The project’s main focus will be to build eye-screening AI-software, which can detect and grade several eye-diseases and eye pathologies. The secondary focus will be to explore the possibility of detecting diseases (or indications thereof), which are not traditionally detected through analysis of the retinal imagery e.g. Alzheimer’s. Such a system will aid greatly in achieving our mission of enabling accessible and efficient eye-screenings, saving sight and saving lives. The system could be used by general practitioners, primary care sector, private contractors such as optometrists and pharmacies and rudimentary screening camps to screen the general public for sight-threatening or life-threatening diseases (where screening and early treatment makes sense) in a convenient and accessible manner.

Traditionally, such AI-software would need to be “hand-made” or programmatically instructed to find specific indicators of each disease, which is a highly tedious and resource intensive method of implementing AI. Increased processing power and large amounts of data have paved the way for a new generation of “self-learning” AI, utilizing deep-learning techniques, which allows for a less time consuming and less resource hungry method of training artificial intelligence models. Artificial intelligence, powered by Deep Learning, is expected to change the health care sector significantly over the next decade.

Related Content