Highly successful Generative AI x Healthcare Hackathon.
The UKRI Centre for Doctoral Training in AI for Healthcare (AI4Health CDT) at the Department of Computing recently organised a highly successful Generative AI x Healthcare Hackathon, showcasing a range of novel prototypes at the intersection of foundational AI and HealthTech. The 2-day event drew 70+ enthusiastic participants from all over the UK and was generously sponsored by leading organisations, including OpenAI, Microsoft for Startups, and Hummingbird Ventures. Powered by OpenAI API credits, the event offered a unique opportunity for participants to apply generative AI technologies to problems within the AI4Health CDT’s main research themes.
The Hackathon aimed to foster collaboration between different disciplines in the HealthTech space, bringing together teams from diverse backgrounds, including clinicians, machine learning scientists, software engineers, UX designers, and regulatory specialists. The projects pitched by the participating teams spanned a wide range of applications; from privacy-preserving health data platforms to educational tools for radiology applications, reflecting the diverse expertise and creativity of the participants.
The hackathon drew over 70 participants from around the UK including clinicians, machine learning scientists, software engineers, UX designers, and regulatory specialists.
Team AutoDeclare brainstorming on how generative AI can help accelerate the regulatory pathways for medical technologies.
Team AutoDeclare brainstorming on how generative AI can help accelerate the regulatory pathways for medical technologies.
Team CuraWell practicing their pitch for an all-in-one privacy preserving health data platform.
Team Heartbeat Guardian pitching their work on a real-time bio-signal monitoring.
Following two days of intensive design and coding, the teams presented their prototypes to a panel comprising both AI and healthcare experts from Hummingbird VC, Vital, Anima, Co:Helm, Humanloop, and Microsoft Research. The winning team, RoboDoc (formerly Paco), developed a prototype innovative digital bridge that aims to connect patients and clinicians for post-surgery recovery. As recognition for their exceptional work, they were awarded the first-place prize of £2,000 in cash from Hummingbird VC and $2,500 of OpenAI credits.
Two runner-up prizes were also awarded to LabGPT for their chat-based molecular biology assistant and Raytwelve for their solution to ground large language models (LLMs) to address healthcare misinformation. All three of these teams have been given further mentorship from Hummingbird VC and Microsoft for Startups to continue building their ideas.
Professor Aldo Faisal, Director of the UKRI Centre in AI for Healthcare, expressed gratitude to the sponsors and participants, saying, "We were delighted that the center's support could catalyse our resources by our amazing partners and we look forward to seeing how the technologies created here continue to develop”. For those interested in learning more about the Generative AI x Healthcare Hackathon, detailed information of the event can be found on the hackathon website and participant technical pack. This Hackathon serves as a testament to the pioneering research and innovation led by the UKRI CDT AI for Healthcare and the Department of Computing at Imperial College London. By fostering collaboration between academia, industry, and the broader community, these institutions are driving advancements in computer science and its applications in various fields, including healthcare.
Organisers
The hackathon was organised by students and staff of the AI4Health CDT at Imperial College London: Alex Jenkins, Avish Vijayaraghavan, Britta Ross, Ahmed Fetit, Safa Abu-Sway, and Aldo Faisal.
Acknowledgements
The hackathon was financially supported by
- UKRI Centre for Doctoral Training in AI for Healthcare (grant number EP/S023283/1).
- Hummingbird Ventures
- OpenAI
- Imperial College Graduate School
Article text (excluding photos or graphics) © Imperial College London.
Photos and graphics subject to third party copyright used with permission or © Imperial College London.
Reporter
Mr Ahmed Idle
Department of Computing
