There have been a number of services over the years that have allowed us to take photos of symptoms. Originally, these services would then offer a second opinion on those images, but increasingly AI has been used to provide an instant diagnosis for the user.
One of the problems with such services is the huge variance in the quality of photo submitted. The same symptom could look markedly different depending on the lighting or general sharpness of the photo.
Israeli company Healthy.io have come up with an app, called Dip.io that aims to overcome this. It allows a standard smartphone camera to perform clinical-grade urine analysis. The user is required to take their picture against the background of a proprietary color card, with the app then using this to correct the colors so that it appears as though it was taken in a neutral light. The test is then analyzed by the AI in conjunction with the medical history of the user.
Supporting home tests
Dipsticks are used for all manner of things, but whilst pregnancy tests are reasonably accurate, those that require color matching are generally not. The company has already worked with the National Health Service to help people with multiple sclerosis perform urine analysis tests at home. This kind of function is vital as MS sufferers are particularly vulnerable to urinary tract infections, with nearly 10% of sufferers getting severe UTI per year.
Patients currently have to visit a clinic when they believe they have symptoms of a UTI, whereas being able to perform the test at home could not only speed up treatment, but also save the NHS around £10m a year.
The company are working on another form of the app that will provide dipstick tests for chronic kidney disease. There are believed to be 26 million or so people with the condition in America alone. An early detection of the disease can avoid expensive dialysis. The urine tests can monitor for protein, with people then given blood pressure drugs to slow the progress of the disease.
It’s a fascinating approach with a wide range of possibilities. Indeed, the company themselves believe that areas such as dermatology could also benefit from their image detection technology. The ability to diagnose skin conditions from a selfie, or indeed to monitor the progress of conditions over time, could be hugely valuable.
With the technology built into phones improving at a rapid pace, the possibilities really are significant. The company believe that they may even be able to offer spectroscopy before too long, or even use infrared and ultraviolet parts of the spectrum to analyze wounds and surface infections.
It’s a fascinating area, and Healthy.io are at the forefront of it. Both will be well worth tracking with interest in the coming years. Check out the video below for more information.