Google’s DeepMind division has announced a partnership with the NHS’s Moorfields Eye Hospital to apply machine learning to spot common eye diseases earlier. The five-year research project will draw on one million anonymous eye scans which are held on Moorfields’ patient database, with the aim to speed up the complex and time-consuming process of analysing eye scans.
The hope is that this will allow diagnoses of common causes of sight loss, like diabetic retinopathy and age-related macular degeneration, to be spotted more rapidly and hence be treated more effectively. For example, Google says that up to 98 percent of sight loss resulting from diabetes can be prevented by early detection and treatment.
Two million people are already living with sight loss in the UK, of whom around 360,000 are registered as blind or partially-sighted. Google quotes estimates that the number of people suffering from sight loss in the UK will double by 2050. Improvements in detection and treatment would therefore have a major impact on the quality of life for large numbers of people in the UK and around the world.
The new project will be working with two kinds of eye scans: traditional fundus images, which are essentially photographs of the retina at the back of the eye, and optical coherence tomography (OCT) scans, which are a cross-section of the retina, giving eye health professionals a more detailed picture of any damage.
However, as the DeepMind site explains: “OCT scans are highly complex and require specialised training for doctors and other eye health professionals to analyse. As a result, there are often significant delays in how quickly patients can be seen to discuss their diagnosis and treatment. To date, traditional computer analysis tools have been unable to solve this problem.”
As Ars reported back in February, Google’s DeepMind group is already working with the NHS to develop patient care software. However, the latest announcement seems to be the first time DeepMind has started to apply its core machine-learning skills to produce medical diagnoses. Interestingly, in May Google announced another machine-learning project that aims to improve early detection of diabetic retinopathy in eye scans. A Google spokesperson told Ars that “this is a completely separate project,” and noted that detecting retinopathy “is a good application of machine vision technology which a few different teams at Google work on.”
DeepMind’s previous collaboration with the NHS became mired in controversy when it turned out that the project was being granted access to the full medical history of 1.6 million NHS patients—who were not being made aware of that fact. The new project seems unlikely to provoke similar fears, since the data is completely anonymous.
Google explained: “It’s not possible to identify any individual patients from the scans. They’re also historic scans, meaning that while the results of our research may be used to improve future care, they won’t affect the care any patient receives today. The data used in this research is not personally identifiable.” Since there is no way for researchers to identify individual patients, “explicit consent from patients for their data to be used in this way is not required,” the company said. In addition, all data will be destroyed at the end of the project.
Throughout the project, the database remains the property of Moorfields. However, “artificial intelligence software developed by DeepMind during the research project is owned by DeepMind and a licence is granted to use the data provided, only to the extent it is incorporated into any software developed from the research.”
This post originated on Ars Technica UK