Big data is helping to treat incurable and rare diseases by speeding up diagnosis

Big Data

The endless reams of data that have inundated businesses on a daily basis for years is more useful than ever thanks to recent advances in data centre technology. Something that would once be considered insurmountable, can now be analysed computationally to reveal patterns, trends, or associations to give invaluable insights, leading to better decisions and solving real-life business problems.

We call this technology big data, and it has helped numerous industries improve and evolve in recent years, with the health industry being one of them. As one of the fastest growing fields, healthcare has had to adopt big data in order to organise the sheer volume of patient histories, diagnosis charts, and vital medical information, and making access to these available within seconds.

The exciting news is, we are only just seeing the potential of big data in the healthcare industry. In the UK, huge anonymised datasets are being developed for areas such as pharmaceutical research, with the aim of improving the efficacy of drugs. Disease research is also getting a big data injection to help tackle conditions such as cancer and diabetes.

Not only will this burgeoning innovation help doctors better understand patients’ medical histories but according to a report by consulting firm McKinsey & Co, it will cut down on the nearly 40,000 deaths a year that stem from misdiagnoses. Big data will also help medical professionals focus on innovation and improvement for healthcare, saving the industry hundreds of billions worldover and anticipating health issues before they become develop into more serious issues.

Harnessing big data with AI

One-time PC maker and now science research company IBM is one of the biggest players in the digital healthcare scene when it comes to big data. The firm has been training its super-smart artificial intelligence (AI) platform Watson over the past few years to harness the power of big data and work with doctors as part of an experiment to solve some complex medical cases. Using AI, IBM can diagnose extremely rare ailments that would normally take doctors weeks or even months of deep probing patients’ backgrounds - in seconds.

Watson, which is famously known for winning the game show Jeopardy! in 2011 after being trained on 200 million pages of general knowledge data, was recently piloted at a private hospital in Germany to work alongside doctors.

Kicking off in December last year, the initiative was made possible thanks to IBM’s partnership with private hospital group Rhon-Klinikum, and was integrated in the form of a Watson-powered cognitive assistance system. The doctors at the Undiagnosed and Rare Diseases Centre in the University Hospital Marburg have trained it on peer-reviewed rare disease literature to help them spot rare disease and make decisions more quickly and safely for hundreds of rare-disease patients, annually.

Watson’s cognitive technology will process the vast amounts of big data that doctors usually have to sift through. Sometimes, the smallest detail can lead to a diagnosis and Watson will be a diagnostic assistant to help doctors solve the case. For Watson to work efficiently, patients fill out questionnaires to give the supercomputer a view of their medical history that might reveal potential exposure to diseases.

But IBM’s Design Principal and executive creative director, Gorham Palmer, explained that making earlier diagnosis it not as simple as just giving the Watson computer access to big data.

“It can interpret images such as CT scans for MRIs and look across the millions of images ingested and actually help make determinations,” he said.

And for those worried about AI replacing humans and taking the social element out of healthcare, Palmer reassuringly added that Watson has been designed to work alongside doctors and assist in their everyday responsibilities, such as performing check-ups, diagnosing patients, prescribing medicine, and understanding behavior modification.

"Watson is not about replacing people, it's about helping people make better decisions,” he said. “It's giving people access to the information they need in a way that's easily accessible to actually help them make decisions.”

So far, the Watson computer in the German hospital has looked at half-a-dozen cases, but it is unclear how many it has correctly diagnosed. However, now it’s been proven to work, the system will be installed and based at the University Hospital of Marburg’s Centre for Undiagnosed and Rare Diseases Centre to help with doctors day-to-day diagnostics.

Gaming for good

Another project that is successfully using big data to compact disease through earlier diagnosis comes in the form of an app called Sea Hero Quest, an initiative by Deutsche Telekom.

Launched in spring 2016, the app aims to gather better research in the fight against dementia, of which there is currently no cure. According the Global Alzheimer’s Report 2015, over 47 million people are living with dementia worldwide, and scientists still know very little about what causes it. But what they do know is that one of the first effects of the condition is a loss of navigational skills.

Realising that scientists need to find out more about how the human brain works while navigating in order to identify what goes wrong during the onset of dementia, Deutsche Telekom looked for a way to improve research by harnessing the power of big data in its network and services.

"Dementia attacks your brain and your brain cells die, and when it eventually kills you, you've lost the weight of an orange’s worth of your brain,” explains Deutsche Telekom’s associate director, Lucy Boyd.

“But obviously the process starts much earlier than someone who has Alzheimer's or dementia who you’d see on TV for [instance]. It's very subtle and it comes on very slowly and what we are not able to do at the moment is use diagnostic tools to get it early enough, which gives the drugs a chance to work.”

At the moment, researchers are unable to diagnose dementia early enough, and therefore when patients start drug trials, they are starting when it's too late to have an impact, Boyd explains.

“And that's why we need to find new ways of researching and putting time and effort into projects like this because we can find new and innovative ways to advance research that aren't just about throwing money at the issue,” she adds. “It's using new mobile technology to see how we can develop new more successful diagnostic tests that give the drugs a chance to work."

Sea Hero Quest was born in the form of a mobile game that anyone can play, to help generate the largest ever crowd-sourced global benchmark for how humans navigate. The database created by the app helps identify the brain’s capacity for orientation and spatial awareness, on a mass level. As you navigate your way through each level of the game, information about how you navigate is collected, without interrupting your game play in any way. At the end of each session, this data is anonymised and sent to a scientific database.

Before the app came along, the largest study ever conducted into human spatial navigation comprised only 599 volunteers. The constraints of conducting research studies in the lab make it very difficult and time consuming to study groups much bigger than this. But by creating a game on a mobile platform, this study has been opened up to people all over the world. It would take over five hours of traditional research to gather the same amount of data as one person playing Sea Hero Quest for just two minutes.

Boyd wants to ensure that people realise Sea Hero Quest is not a test for dementia, nor is it designed to be played by those who have the condition. It has been designed for people of all ages to play, providing scientists with data on how the average human brain's ability to navigate.

The development of such a benchmark is widely acknowledged as one of the key steps towards developing new diagnostic tests for the diseases that cause dementia. And it is only by creating this rich population-level understanding of human navigational abilities that scientists will be able to effectively compare and detect the problems experienced by people with dementia.

It’s about collaboration

Without the relevant and appropriate information at their fingertips, doctors are limited in their ability to provide patients with the correct diagnoses. But with the use of big data, they are not only able to organise patient information; integrate and balance consideration of patient symptoms, history, and environmental factors, but also be given more insightful data that can make their jobs as life-savers much easier.

And as electronics giant Philips’ chief health officer, Jan Kimpen, notes, big data won’t be a matter of choice in the future.

“We will have to make use of data of all forms to make the best of healthcare,” Kimpen said. “And we will have to do it it's not a choice to do it. Data is absolutely an important aspect of future health.”

Kimpen is deeply convinced not one company in the ecosystem can do it alone; for big data to make the biggest and most valuable impact in the future, he believes the idea of organisations (big and small) working together is important for “the best solutions for the patients.

He added: “We need to start looking at a number of partners, hospital partners, insurance companies, etc, and starting conversations with pharma companies to make personal medicine come to life.”

Find out more

Get The Drum Newsletter

Build your marketing knowledge by choosing from daily news bulletins or a weekly special.