Dan Vahdat was fast asleep at the back of a Royal Airforce plane headed for China, when Britain’s prime minister Theresa May started walking up the aisle towards him.
When she was about six feet away and chatting to a group of passengers nearby, the entrepreneur woke up with a start.
“I thought, I better get ready,” he remembers.
Vahdat, who is CEO of British medtech startup Medopad, was part of a 50-strong trade delegation accompanying May to Beijing and Shanghai last week.
For the politicians it was a way to build stronger ties outside Europe in preparation for Brexit. For Vahdat it pinned down partnerships with some of China’s biggest companies, paving the way for clinical trials that could make his software better at predicting ill health.
It also meant getting a selfie just in time with the Prime Minister:
Vahdat’s company, founded in 2011, makes software that doctors and nurses use to track the vital signs of patients they’re treating for cancer or chronic diseases.
The app can collect metrics like heart rate data from a Fitbit, or blood-sugar levels from a glucose monitor.
It's crossed over into general health monitoring by insurance companies like China's Ping An, and Volkswagen plans to use Medopad to monitor the health of more than 100 of its truck drivers.
The startup recently closed the first part of a $120 million financing round, getting $28 million from a group of investors led by China’s NWS Holdings.
Like competitors uMotif in London or New York’s LifeOnKey, Medopad's secret sauce is in predictive analytics.
That is, algorithms that can draw inferences from a patient’s metrics to suggest a diagnosis to a doctor.
One way to make those algorithms smarter, Vahdat recently realized, was going to China and collecting health data on large swathes of the country’s population.
“We started going there nearly four years ago,” he says. “We wanted to get there simply because of the scale of numbers. If you have scale you have access to an asset that nobody has.”
Data analytics is fundamental to machine learning, and a larger data set can lead to smarter predictive capabilities, says Vahdat. “It gives you an edge against other companies in terms of the predictions you can do.”
Hence Medopad is now working on two large medical studies in China.
Vahdat won’t discuss details of the first study with Chinese Internet giant Tencent, though he mentions that Medopad can get access to “1 billion users” through its popular messaging app WeChat.
The second is with the Capital Medical University’s Geriatric Center in Beijing, where Medopad will track the walking gaits of patients with Parkinsons. That partnership gives the startup access to up to 1 million patients, says Vahdat.
Medopad has partnerships outside of China too, with institutions like the NHS in the UK and a more recent deal to develop predictive software for doctors with John Hopkins University.
But the sheer scale of people grouped together under institutions like hospitals, or chat apps in China is unprescendented.
“Depending what kind of studies we want to run we can target 10,000 or 100,000 [people],” Vahdat adds. “I don’t think anyone can access 100,000 outside of China.”
