Hacking Health (Without a Scalpel)

By Dejan Jotanovic & Scott Ritchie

This year’s #HealthHack had two sites, rainy Melbourne & (brand new) sunny Sydney, and was on the operation table for three days: October 24th - 26th. But let’s give a quick diagnosis - what IS #HealthHack? As explained on their website:

HealthHack is a weekend data hack dedicated to problems that medical researchers face. Scientists often create vast amounts of data, but don’t always have good ways to analyse, visualise and communicate it. This weekend will bring together software developers, user experience designers, data analysts and visualisers working directly with researchers to create new, better tools.

Basically a bunch of smart people are kept in a big room (cheers, ThoughtWorks) with a load of dumplings, sticky-notes and coffee coupons. Teams are formed, questions are asked, data is tugged, pulled, sculpted. Two sunsets later and the answers are presented in the form of apps, videos, visualisations and speeches. What you can do with the right people in the span of two and a half days is truly outstanding. The event was brought to life by the Open Knowledge Foundation.

Here’s a quick recap of the different Melbourne teams and their collective projects.

  1. Epilepsy Visualisation Tool

Kate Birch and Anne Mcintosh, two researchers from the University of Melbourne,  wanted a tool that could be used by clinicians to visualise patient data, so that they can more easily  find correlations between seizure frequency, medication, quality of life, and any potential surgery.

Why? Because epilepsy is a very complicated disorder: seizures frequency and severity vary from person to person, having different impacts on each individual.  Understanding how patients are impacted by epilepsy is further complicated by the messy way in which patient history information is stored.

A team of four created a tool allowing the researchers to visualise patients’ seizure history. This could help clinicians make better, more informed decisions. And it was mobile friendly!

You can read more about this team here.

  1. Epilepsy Management Team

Same problem. Another solution! 

This team created visualisations which would help clinicians quickly and easily see their patients seizure, surgery and medication history! It’ll also how any medication frequency changes or how severe the seizure experiences were.

Read more here.

  1. This is not a pipe: VLSCI Bioinformatics Pipeline

Let’s all be honest here, sequencing and processing genomes can be tough. Outputs, inputs, the creation of pipelines. Often very complicated pipelines! Dr Bernie Pope and Dr Phillipa Griffin from VLSCI wanted to improve Rubra, a pipeline management tool used at VLSCI, to make it more user friendly for the biologists doing the sequencing, since they often don’t have a lot of computational experience!

The team made some serious progress into improving Rubra - and here’s hoping they continue!

Here are all the details.

  1. Mutation Instability Filter (MIF)

You know how it goes: you’re trying to identify genetic mutations responsible for a disease within a family, so you sequence their DNA, and after a series of analysis and filtering steps you’re left with a couple of hundred potential genes. Identifying the correct one is usually expensive and time consuming, requiring many experiments in the lab. But what if we could speed up this process AND reduce the cost? Enter, MIF.

Dr Charles Galea wanted to create an automated pipeline to query protein databases, automatically matching genes to their proteins, and measuring whether the mutations identified significantly disrupt the protein structure. This would hopefully allow researchers to massively narrow down the list of candidate genes from a few hundred to just a handful to follow up on. 

A team of developers, scientists, and user experience designers were able to successfully build this pipeline over the weekend, creating a filter that dramatically decreases the cost and time required to identify the genes underlying familial diseases! This is the bit where someone in the audience would typically shout “That’s so awesome”.

And that they did. MIF then went on to win the coveted “Spirit of #HealthHack” award, for being so damn awesome.

Check out all the awesome details here.

  1. GIRROR: Tracking your emotions and gambling behaviour

How do environmental and emotional factors impact addictive behaviours like gambling? That’s the question Ben Fulcher and George Youssef from Monash University wanted help answering. To do this, they wanted to create a smartphone app, to allow gambling addicts to report on the factors that might influence their decision making.

They cleverly created an app (available for both iOS & Android - score!) which used pop-up messages asking patients to complete questions about their emotional state and environmental context. A short reaction time game was also included.

Sounding cool? Well, it was! The future of this project lies in the ability for practitioners to develop personalised treatment plans, targeting the special things about patients - their wide ranging emotions and environmental contexts!

It’s therefore no surprise that team GIRROR came second at the #HealthHack awards! Read all about the project here.

  1. VizMyGrant

Every year medical researchers across Australia eagerly await the results of the NHMRC funding, which are announced in late October. It’s an exciting and terrifying time for everyone: will they receive funding this year to pursue their research questions? Or will they have to join someone else’s lab? Or even contemplate moving overseas? Although the NHMRC makes their data publicly available, you have to pore through spreadsheets, pdfs, and word documents if you want to go any deeper than the summary statistics the NHMRC provides. Dr Marguerite (Maggie) Evans-Galea wanted to solve this problem by creating a tool to generate user-friendly graphics to clearly explain where the funding goes each year. Maggie was particularly interested in visualising the gender balance in funding outcomes at different career stages, having spent many hours trawling through the documents several years ago to visualise those imbalances.

But what if we could take it a step further? What if we could create an interactive tool that dynamically creates user-friendly graphics to view the NHMRC funding landscape in any way an interested individual could think of? What trends might that reveal? Our very own Scott Ritchie joined team VizMyGrant with the aim of creating an interactive web-based tool for visualising the relationships in the NHMRC funding outcome data.

 Scott, along with the rest of team VizMyGrant spent the weekend cleaning up the NHMRC funding outcome data and creating a suite of tools and visualisations, all of which were deployed online live at Health Hack! VizMyGrant proceeded to make waves on twitterverse with their visualisations of the gender imbalances in funding outcomes:

With the potential impacts on policy, and broad utility for early career researchers, its no surprise that Team VizMyGrant went on to win the Melbourne #HealthHack awards. 

You can find out more about Team VizMyGrant by watching their video, visiting their team page, or following them on twitter @VizMyGrant. All of the interactive visualisations created by the team can be viewed online, and Scott Ritchie’s tool can be found at https://viz-my-grant.shinyapps.io/view.

—–

Want to learn more about #HealthHack? Read organiser Maia’s musing here, and Tim Hildred’s (from Open Source) here.  And check out the official storify.

Want to plan an event like #HealthHack and join the Open Science movement? Get into contact with the Open Knowledge Foundation. Ping them on Twitter or check their website.