Artificial intelligence to diagnose eye diseases
23rd February 2018 – Scientists say they've developed artificial intelligence (AI) which can be used to swiftly and accurately diagnose serious eye conditions.
In the latest issue of the biology journal Cell, they describe a way of using artificial intelligence and machine learning to recognise macular degeneration and diabetic macular oedema, the two most common causes of irreversible blindness.
They say this new technology could be used anywhere in the world and could prove especially important in rural areas, and places like China, India, and Africa.
AI and machine learning
Traditional computer programs work by defining actions to take given certain circumstances - such as typing in some numbers and doing calculations. Type in something it isn’t programmed to react to - like a word instead of a number - and it won’t know what to do.
AI is a part of computer science where machines are designed to have some intelligent behaviour.
Machine learning means getting systems to do things without being explicitly programmed to do so.
Like a child learning something new the techniques involve human experts letting the system know whether a decision it took was right or wrong - and the machine learns from this to identify images, patterns, behaviours, and more.
This technology is already used in speech recognition, facial recognition, and self-driving cars. The hope with this medical application is that machines could become as expert as doctors - but quicker - and able to work around the clock - in identifying signs of eye diseases.
Diagnosis and treatment
Not everyone is willing to trust artificial intelligence with something as important as their health but the team of scientists from the US and China who came up with the new technology are hoping to change minds.
Prompt treatment is important for macular degeneration and diabetic macular oedema but currently patients suspected of having a retinal disease face a lengthy referral process before being seen by a specialist for treatment.
The scientists believe having an automated diagnosis performed by AI could mean patients see a specialist and get treated sooner. In some cases, they say, this could be the difference between patients seeing or losing their sight forever.
Research and findings
For a study, researchers at Shiley Eye Institute at the University of California (UC) San Diego Health and UC San Diego School of Medicine, with colleagues in China, Germany and Texas, arranged for the new AI technology to look at more than 200,000 optical coherence tomography (OCT) images of people's retinas. The researchers then compared the diagnoses from the computer with those from 6 ophthalmologists who also reviewed the scans.
They say with simple training, the machine was able to perform to the level of a well-trained ophthalmologist. It could also generate a decision on whether the patient needed to be referred for treatment within 30 seconds and with more than 95% accuracy.
The researchers say the new technology can also distinguish between bacterial and viral pneumonia in children based on chest X-rays and it has other potential applications. Consequently, they have published their information and data open-sourced so that others can further improve, refine and develop it.