OF THE millions of photos shared online every day, which most faithfully represent their subjects? The popular #nofilter hashtag would suggest it is those that have not been digitally altered. But photographs of the same thing can differ greatly, depending on ambient light and the distance and angle they were taken from. So the right manipulation can actually make a picture more honest—and therefore more useful for medical purposes.

That is the idea behind an app from Healthy.io, an Israeli firm. Dip.io, as this app is known, uses mobile-phone cameras for clinical-grade urine analysis. The patient follows the instructions, waits for the colours on the dipstick to develop and then takes a picture of it against the background of a proprietary colour card. The app uses the card to correct the colours so that the dipstick appears as if in a neutral, standard ambient light. The result is then analysed automatically, in light of the patient’s medical history. If this analysis suggests a consultation or prescription is needed, that can also be arranged automatically.

The first urine-dipstick test was developed in 1956, to look for glucose, which indicates diabetes. Since then, sticks have been used to test for the presence of blood, of protein, of hormones indicating pregnancy and also of various bacteria that cause urinary-tract infections. Some sticks, notably those employed for pregnancy tests, can be bought over the counter and used at home. But for tests that require colour-matching, rather than merely checking whether a single line is present or absent, home analysis is regarded as unreliable.

Point, click, treat

It is this unreliability that Healthy.io is attempting to deal with. The firm is, for instance, working with doctors in Israel on a system that lets pregnant women at risk of pre-eclampsia (dangerously high blood pressure, which is signalled by protein in the urine) use dip.io to monitor themselves at home. In Britain, meanwhile, the National Health Service (NHS) is starting to employ a version of the app to monitor those suffering from multiple sclerosis whose bladders are affected by the disease. Members of this group, which is around 60,000-strong, are at particular risk of urinary-tract infections. About 5,000 of them develop severe infections every year. At the moment, when someone in this position spots early symptoms he must go to a clinic to be tested. Home-testing, followed by a prescription posted to those who need it, should obviate that need, speed up treatment and also save the NHS around £10m ($12m) a year.

A third dipstick test the app may soon be applied to is chronic kidney disease. In America alone some 26m people have this condition, which is often associated with diabetes and high blood pressure in a phenomenon known as metabolic syndrome. Patients in the late stages of kidney disease need costly dialysis. But if the illness is detected early, by screening the urine of those at risk to check for protein, sufferers can be given drugs that lower their blood pressure and thus slow the disease’s progress.

Nor is urine analysis the only part of medical practice that may benefit from healthy.io’s standardised selfies. Dermatology should profit, too. To diagnose a skin condition from a picture, or to monitor its development over the course of time, dermatologists need not only to control the colour of an image, but also its size and the angle from which it is taken. In this case Healthy.io’s answer involves a sticking plaster printed with coloured hexagons that is placed near the relevant patch of skin. Like the dipstick card, the plaster acts as a reference which the app uses to correct and standardise the resulting image.

Yonatan Adiri, Healthy.io’s founder, has ambitions beyond even this. As he observes, phones are everywhere and are improving all the time without his firm having to lift a finger. By using their built-in cameras, the company can piggyback on phone-makers’ research and development. Soon, his app may be able to employ spectroscopy—a detailed analysis of the frequencies of light making up an image—or extend its range beyond visible light into the infrared and ultraviolet parts of the spectrum. This may help analyse wounds and surface infections, by studying characteristics that are invisible to the naked eye. That will save both doctors and patients time, and should result in better outcomes all round.