Tim Thorne

Leaders Changing the World no 3: Tim Thorne

Tim Thorne and I worked together at RBS to create the Group Innovation function. He was Innovation Director and I was the part-time Chairman of the Innovation Board to which Tim reported. It wasn't the most promising environment for new ideas to flourish but Tim was a leader with an unstoppable sense of mission and accomplished a lot - nothing capable of changing the world though.

But now he is changing the world. He is up to a lot of stuff , but it's a project he is leading at Cancer Research UK which I want to write about. The project is finding a way to speed the development of new cures for cancers - this is real ground-breaking innovation which is capable of changing the world for a lot of people!

Tim had a look at the end to end research process to see how it could be speeded-up - it takes years to get a new treatment or diagnostic from the lab through trials and into widespread use. One of the things holding it up is data analysis. Do you know that it can sometimes take 3-4 years to complete the analysis of the results of a drug trial or that it involves analysing petabytes of DNA data to develop new diagnostic tests yet the average human genome is the data equivalent of a 20,000 page phonebook! Much of this analysis is boring and repetitive, yet around the world it is undertaken in labs by highly qualified and well-paid PhDs.

There has to better, faster and cheaper way of doing this analysis! So here was one of Tim's ideas. NASA is already using the internet to crowdsource the analysis of images from the Hubble telescope - could we use the same platform to crowdsource the analysis of microscopic rather than telescopic images? So Tim's team got the astronomers together with a bunch of young developers and some of the leading cancer scientists in the UK. They locked them in a room at the Science Museum for a long weekend.

In short they quickly developed a prototype and within 3 months they launched "cellslider" on the internet

www.cellslider.net

The results have been amazing. In under 6 months they have had an astounding 1,000,000+ images classified - the same data would take a team of 10 pathologists, giving a few hours a week above their day-to-day NHS work, 3.5 years to process! They spent only a few thousand pounds on marketing it - so costs were next to nothing! What's more they will shortly publish findings that prove it is just as accurate. So it's proved the concept that crowdsourcing can take years off the time to analyse data and speed new cancer treatments and diagnostics to market. By giving a few minutes of your time you can really help save thousands more lives every year!

Well things have now gone really crazy from there!

- they are recruiting users from all over the world - from a small class in an Alaskan High School to thousands of Chinese after a post on Reddit.
- They took it Parliament and 72 MPs have become users.
- Amazon love it and have given free use of Amazon Web Services hosting
- Google love it, are regularly asking Tim's team to present at their gigs
- Facebook love it - posts on it achieve record click-through rates, they have given free developers and appeared with Amy Carton from Tim's team on Radio 4′s Today programme
- They are being pro-actively contacted by potential corporate sponsors, business figures, foundations and philanthropists who want to help.
- Even MIT loves it and has a group of students helping out

They have now moved on to look at a different type of data - the DNA microarrays which can indicate where the mutations occur that are responsible for cancers. This data is really boring to look at - endless streams of dots on a graph. Too boring to engage users. So instead they are going down the route of creating smartphone games that are powered by these data streams. For example the data becomes asteroids for your spaceship to avoid or drops of rain to collect in a bucket - your movements in reaction to the game then do the analysis for them. Last month they got together developers from Facebook, Amazon, Google and leading games companies for another weekend. Together they created 12 working prototypes of games which we can now launch later in the year. This was the first time these companies had ever put their developers together in a collaboration and as a result there was huge media interest in the weekend.

So where now?

They are looking at numerous different ways to bring big bucks into this field - from conventional grants to creating a business model which would allow VCs to invest.

They will build a new platform .The platform would bring in various data-sets, do all the back end stuff and then throw out APIs into applications and games. Some of these might be more Cancer Research micro-sites but they would also have open APis and allow others to develop games and apps using their data. They could also take in data sets from outside Cancer Research UK - there are many other medical research and genome projects collapsing under the weight of the data they need to analyse.

This could really change the world!