Could digital detectives solve an ancient puzzle?
- Published
For more than two thousand years people have believed that joint pain could be triggered by bad weather, but the link has never been proven.
Now, by harnessing the power of thousands of volunteers, doctors hope to unravel the mystery. And the new technique could offer countless solutions to a whole host of ailments.
"I'm always in pain, 24/7," says Becky Mason, sitting at home on her sofa in Alsager near Manchester.
Like millions of people around the world she suffers from pains in her muscles and stiffness in her joints.
"I know, if it's going to be a very damp cold day, it's likely that my pain is going to be worse."
She has discussed it with her GP and has always wondered if there really is a link between her pain and the weather.
Becky isn't alone. The link between joint pain and bad weather has long been suspected by patients and medical professionals alike and the theory dates back at least to Roman times and possible earlier.
"Is it an old wives tale? Am I imagining it?" she asks.
It's a question she finally hopes to answer, not by visiting a hospital or undergoing tests, but simply by using her smartphone.
Each day she enters information about how she feels into an app on her phone, the phone's GPS pinpoints her location, pulls the latest weather information from the internet, and fires a package of data to a team of researchers.
On its own Becky's data is of limited interest, but she isn't acting alone. More than 13,000 volunteers have signed up for the same study, sending vast quantities of information into a database - more than four million data points so far.
The app, called "Cloudy with a Chance of Pain" is part of a research project being run by Will Dixon. He is a consultant rheumatologist at Salford Royal Hospital and has spent years researching joint pain.
"At almost every clinic I run, one or more patients will tell me that their joint pains are better or worse because of the weather" he says, but until now he has never had the means of collecting enough data to find a conclusive answer.
Which is perhaps a good point to explain Will Dixon's other job title - Professor of Digital Epidemiology.
Traditional epidemiologists study health and disease in particular populations. Usually it means collecting data in person - asking patients to visit you, or heading out into the field. 'Shoe leather epidemiology', it is sometimes called.
But digital epidemiology allows patients to send detailed information over the internet - which means they can do it more regularly, and of course you can get many more people to take part, thousands more; numbers that would be unthinkable using the old methods.
By combing through that data, Professor Dixon hopes it will be possible to find correlations and clues that would have been hidden to doctors just a decade ago. His team will analyse the data over the coming year, and hope to find a definitive answer to the question.
World Hacks is a new BBC team looking at global problems.
We meet the people fixing the world.
The technique isn't just limited to arthritis research.
Another study underway in the US has recruited more than 20,000 participants using an app that asks them to say "ahhhhhhh" into their phone.
Named mPower, and built using technology developed by British academic Max Little, the project hopes to find out more about the onset and progression of Parkinson's disease. If the "ahhhhhhh" sound is smooth and unbroken, it has likely come from a healthy patient. But if it breaks and wavers, it could suggest that the patient may have Parkinson's.
By monitoring the precise pattern and pitch of the noise, it may even be possible to determine how advanced the disease has become, or how strongly its symptoms are being felt at a given moment. Using that information, it could allow patients to take much more specific doses of a drug to help manage the disease. The software is even being used in a clinical trial for a new drug.
And again, it is the accumulation of vast amounts of data, volunteered by thousands of participants, that is making the study possible.
Another app, soon to be launched, will allow users to photograph their plate of food, and use artificial intelligence to work out what's on the plate. The technology could help people determine the nutritional content of their meal, and allow public health bodies to track how well any particular population is eating.
It is being developed by Marcel Salathe, also a Professor of Digital Epidemiology and founder of what is likely the world's first lab dedicated to the field of study.
He thinks the discipline could have particular benefits in parts of the world where basic medical infrastructure is lacking, but lots of people have smart-phones. Digital epidemiology could become the reporting network through which sickness outbreaks are initially detected, he says.
But vast amounts of data don't come without their own unique set of difficulties, he warns.
"The data can be extremely noisy," he explains. "Dealing with very large data sets and finding a needle in the haystack is very challenging from a technical perspective."
Perhaps the most interesting part of this new technique is the motivation of the people donating their data.
'Cloudy with a Chance of Pain' may never reap rewards for Becky herself, yet she seems quite happy to spend her time putting her data into a smartphone app and then sending this off to a remote location.
"When you're in pain all the time, it's easy to get low," she says "I'm at home and I can't work which makes me feel useless. But [with this app] I can still be helpful, and that's so powerful in my tiny little world, it helps me in a massive way."
Listen to BBC World Hacks on the World Service or listen back on the iPlayer.
- Published31 October 2016
- Published9 January 2017