Behind the Scenes at the Visual Computing Hackathon

The Visual Computing Hackathon, or VIC-HACK 2024, took place in Abacws on 12-14 June.

Organised by the Cardiff University Brain Research Imaging Centre (CUBRIC), the goal of VIC-HACK 2024 was to bring together researchers from a wide range of backgrounds to collaborate on open science projects in neuroimaging and computer science.

Participants came from across the UK and from overseas for the three-day-long hackathon, where projects included:

• Adapting to patient motion in ultrahigh-field MRI
• Automatically labelling medical data from diagnosis reports
• ‘NiiView’: Medical Image Visualisation App
• ‘Serial Scope’: Making an easy way to view real-time sensor data

In Cardiff University, we host a range of hackathons every year, each focusing on a different area or specific issue.

In general, these hackathons bring together groups of computer scientists and other like-minded individuals to creatively engineer solutions over a specific time period.

"It's a very informal event that gathers together people from any background with the common interest of tackling specific problems," said Dr Marco Palombo, one of the VIC-HACK's organisers.

What does the Visual Computing Hackathon involve?

The organisers of VIC-HACK 2024, all based in CUBRIC, are Dr Paddy Slator (School of Computer Science and Informatics), Dr Marco Palombo (School of Psychology), Dr Maëliss Jallais (School of Psychology) and Lewis Kitchingman (School of Psychology).

CUBRIC is a multi-disciplinary group that brings together expertise in brain imaging, mapping and stimulation to better understand the causes of neurological and psychiatric conditions.

Reflecting the work of the Centre, VIC-HACK’s focus revolves around developing solutions to medical imaging and diagnosis in neuroscience.

"We've got a variety of projects from across visual computing , and in particular we've got quite a lot in the field of medical image computing , which I think is really good because that shows how  computer science can be used to have real world impacts," said Dr Slator.

One of the project teams worked on adapting to patient motion in ultrahigh-field magnetic resonance imaging (MRI).

"I work on MRI and when we have patient motion the data becomes inconsistent, so we have to reacquire the data," said Professor Emre Kopanoglu, who led the project team.

Professor Kopanoglu, who is based in the School of Psychology, added: "Correcting for patient motion takes a bit of time, so we cannot do it in real time. But, a lot of patient populations are susceptible to not being able to stay still throughout the scan, so we tried to develop some machine learning methods to estimate the effect of motion on ultrahigh field MRI and correct it on the go."