Clinical Data Hackathon Takes on Real-World Challenges

June 8, 2026
3 min read

Clinical data hackathon 26 1

AI is intertwined in virtually every part of daily life. In healthcare, rising patient volumes, an aging population and increasingly complex medical data have brought the benefits of AI to the forefront.

As a learning tool, generative AI is transforming healthcare by providing data-driven knowledge that improves patient care, enhances diagnostic accuracy and makes clinical workflow more efficient, all of which only increase the demand.  

To address this, the Stony Brook Clinical Data Hackathon brought together 21 clinicians, data scientists and researchers who worked with Google engineers to tackle real-world healthcare challenges using cutting-edge cloud and AI tools. The full-day event took place April 30 at the Medical and Research Translation (MART) building, presented by Stony Brook’s Department of Biomedical Informatics (BMI) and the Institute for Engineering-Driven Medicine, and sponsored by the Department of Anesthesiology.

The event was organized by Janos Hajagos, research assistant professor, chief of data analytics and BMI chief security officer; Jonathon Schwartz, MD, Cardiothoracic Intensive Care Unit (CTICU) critical care attending at Stony Brook Medicine and assistant professor in the Department of Anesthesiology, Renaissance School of Medicine; and Joe Cesaria, BMI chief administrator.

Clinical data hackathon 26 3

“The hackathon was a pilot hackathon designed to work out the format for future clinical data hackathons aimed at bringing folks from across campus together who have varied professional backgrounds and disciplines,” said Cesaria. “For this pilot, the objectives were to help participants gain practical experience with Google Cloud’s Vertex AI, BigQuery, and the Healthcare API, and to collaborate alongside Google engineers to prototype solutions with clinical impact.”

Hajagos added that the clinical data hackathon showcased the impact of collaborating across campus. “Teams made impressive progress on projects ranging from ICU hypotension forecasting and arterial waveform analysis to NYC spatial census visualization tools, operational systems like ‘Rescue Radar’ to improve hospital response times, and AI-driven research on seasonal hospital admissions and ADK development environments,” he said.

The event offered a unique opportunity for Stony Brook clinicians and data scientists to network, create new collaborations and lay the groundwork for future research projects. Participants gained practical experience working with gold-standard clinical datasets, including MIMIC-IV, TCIA and the Imaging Data Commons. From building AI-powered clinical note summarizers to multimodal radiology analysis pipelines, teams collaborated alongside Google engineers to prototype solutions with clinical impact.  

Clinical data hackathon 26 2

Google’s Anand Kumar, senior customer engineer (Applied AI), and Willis Zhang, customer engineer at Google Cloud, led the program and provided guidance to participants throughout the day. 

Participants handled the entire end-to-end lifecycle:  

  • Deep Research: Conducting initial literature and data discovery.
  • Data Engineering: Loading and structuring complex datasets.
  • Model Intelligence: Training custom models and gathering deep insights.
  • App Development: Building full apps and front-end UIs.
  • Communication: Generating final challenge presentations.

“This is not ‘vibe coding,’” said Kumar. “This is a high-level agent executing complex engineering tasks with an expert Google assistant by your side. We’re moving from ‘how to build’ to ‘what to build,’ and the results are incredible.”

“I was impressed with the creativity and technical expertise demonstrated by the participants during the full-day event,” said Hajagos. “The program exceeded all expectations and highlighted the tremendous potential of combining clinical insight with advanced technology to drive meaningful advances in healthcare and research.”

— Robert Emproto