This competition is co-organized by the Cancer Digital Insights Working Group at the Cancer Center at Illinois (CCIL) and the Center for Artificial Intelligence Innovation at the National Center for Supercomputing Applications (NCSA). The hackathon brings together talented University of Illinois students from across disciplines to tackle a real-world cancer research challenge using artificial intelligence, foundation models, machine learning, and advanced computational resources.
(The informational page from the Cancer Center is HERE)
This year’s challenge will focus on applying AI and foundation models to cancer-related pathology and clinical data, giving participants an opportunity to work across the intersection of computing, medicine, and cancer research.
Event Dates
Saturday–Sunday, October 17–18, 2026
Location: National Center for Supercomputing Applications (NCSA)
Teams will work on the challenge throughout the weekend with support from technical and scientific experts. Participants will receive the challenge materials, datasets/resources, technical instructions, and background information approximately one week before the event so they can review the materials and resolve any access issues in advance.
Teams may continue refining their work after the weekend in preparation for flash presentations and judging, planned for Wednesday, October 21.
We encourage you to sign up soon using this Registration Link. If you already have a team be sure to have the team lead complete the form and include the name and contact information for the full team.
Deadline for registration is October 1, 2026.
Eligibility: Teams should have two or more students (undergraduate and/or graduate) with at least one currently enrolled in the Computer Science Department. Students are encouraged to form teams of up to five students and to also include students from biomedical degrees.
Hackathon Project
Multi-modal Pathology Models: Exploring Practical Applications for Improving Clinical Workflows
Science team contact: Charles Blatti (blatti@illinois.edu), Kun Wang (kwang222@illinois.edu)
Technical team contacts: Volodymyr Kindratenko (kindrtnk@illinois.edu) , Josh Allen (jallen17@illinois.edu)
Problem: High-resolution digital imaging of tissue samples has enabled a wave of AI applications in tumor diagnosis, grading, prognosis prediction, and biomarker discovery. Here, we would like to investigate the potential of foundation models specially trained on both pathology images and related medical text to improve the precision, reproducibility, and interpretability of clinical pathology workflows. Pretrained weights for state-of-the-art multi-modal foundation models for computational pathology along with curated datasets of paired images and reports (from TCGA, OpenPath) will be pre-loaded on the NCSA’s GPU cluster for students to quickly access and test their ideas. For example, a team might build a case retrieval system that fetches groups of morphologically related slides in response to a free-text clinical description (e.g., “invasive ductal carcinoma, grade 3, with prominent tumor-infiltrating lymphocytes”), or develop a conversational diagnostic assistant that helps a pathologist avoid ambiguous or uncertain terminology when building their report. Prior work shows benchmark performance of these multi-modal models (CONCH, MUSK, survey). The goal of this hackathon competition is to leverage and augment these models to build innovative and impactful tools for specific clinical applications.
Computing & Technical Resources
Participants will have access to NCSA computing resources, including Delta/DeltaAI, along with instructions for required accounts, credentials, datasets, models, and other technical resources. Technical and scientific experts will be available throughout the hackathon to answer questions and troubleshoot issues.
Judging Criteria
Projects will be evaluated using criteria that include:
- Innovation and creativity in the approach of using AI models, machine learning, and other information science and computational techniques to implement tools and workflows to solve challenging medical and problems.
- Development progress on solution with effective use of computing resources and existing literature and tools, including access to NCSA flagship AI compute platform.
- Relevance and potential clinical impact of the developed solution.
- Quality and clarity of final project summary and oral presentation.
The judging panel will include experts representing both clinical/domain expertise and computational/statistical expertise.
Prize: 1st place $3000, 2nd place $1500, 3rd place $750