TECH

What do your eyes say about your heart? Google's AI may have the answer
Verily, which is one of Alphabet's healthcare technology companies, announced this week that it is working on a number of new projects that could equip medical research labs around the world.In a paper published Monday in the journal Biomedical Engineering, scientists explain that one of the systems is based on an artificial intelligence that, through an algorithm, is able to predict the risk of cardiovascular disease of a patient via ocular analysis. The technology has a database that allows the comparison of results in order to improve their success rate permanently.The article explains that the system takes into account some characteristics of the patient, such as age, gender and blood pressure. The goal is to replace more complex and time-consuming analyzes, such as those currently used for the same purpose.
Verily explains that the project still has some limitations as the system is being tested with 45-degree amplitude images and does not yet have a solid set of examples to be able to operate in real-world situations. The next step in the development process is to densify the number of scenarios in which the technology is being tested, so that it can be used in actual patient testing. To date, the AI has been tested with machine learning systems in the analysis of 300,000 existing medical records, which allowed it to create response patterns.
The anatomical basis for this technological solution lies in the blood vessels in the back of the eyes, which reflect the health status of patients at various levels. By studying the aspect of this part of the eye with chambers and microscopes, doctors will be able to infer various information that are, as a rule, related to the risk of cardiovascular disease.
The results obtained to date are very similar to the quality standard guaranteed by the blood tests, which allow correct screenings in 72% of the occasions. Google's AI is now at 70%, but it has never been tested in real-world scenarios.
To be applied in industry, this technique will allow non-invasive analyzes, cost and time reductions in the screening processes.

Sapo