Our goal is to accelerate data driven discovery via self-service data exploration applications, reliable data models, multimodal datasets and AI enabled data science applications.

Purpose and Scope

The Translational Data Accelerator advances translational research by reducing the technical, operational, and data governance barriers to using sensitive data effectively. Our goal is to accelerate data driven discovery via self-service data exploration applications, reliable data models, multimodal datasets and AI enabled data science applications. We build reusable, scalable, democratized data capabilities and processes to enable staff across clinical trials, clinical research, and laboratory sciences to spend less time overcoming technical or policy barriers and more time advancing science.

The Translational Data Accelerator is responsible for the infrastructure, data products and data governance processes that support translational science, including:

  • The translational data platform including its infrastructure and data access applications (CARDS+)
  • Trusted, governed modeled data products, analytic datasets and research archives
  • Data governance policies, process, stewardship supports, and workflows enabling (re)use of these data
  • AI and LLM enablement via technology and governance supports for responsible use of AI in our work and research

Accelerator History

The Accelerator grew out of the Data Science Lab (DaSL), established and led by Jeff Leek to expand data science capabilities and support data-driven research at Fred Hutch. As this work grew, a dedicated translational data program emerged to address the infrastructure, engineering, data science, governance, and education needed to make clinical and research data more broadly accessible and useful for translational science.

Today, in 2026, the Translational Data Accelerator is becoming an independent program, building on that foundation with an expanded focus on shared translational data infrastructure, research-ready data products, responsible data and AI use, and education and support for the Fred Hutch research community.

CARDS’ History

The Clinical and Research Data System (CARDS) is the Fred Hutchinson Cancer Center’s multi-modal translational data platform — a secure, scalable, and supportable foundation designed to unify and modernize how we manage and use cancer-related care data from clinical and derived research data. CARDS provides a centrally governed yet democratized environment and toolset that enables the Fred Hutch community to build data products and tools leveraging multimodal patient clinical and research data. With CARDS, our researchers, clinicians, and data scientists can accelerate innovation across clinical trials, precision oncology, data science, and AI.

The vision for CARDS was developed following the 2022 merger of Seattle Cancer Care Alliance and Fred Hutchinson Cancer Research Center. CARDS makes it possible to securely integrate data from electronic health records, clinical laboratories, imaging systems, and other sources. This unified ecosystem supports ethical, compliant, and efficient data use — enabling research and operational teams to collaborate effectively while protecting patient privacy and data integrity. The official rollout of the CARDS platform occurred in the summer of 2026.

Our translational data needs that are the focus of CARDS are defined by a broad definition of “cancer-related care”. The Clinical Data Exchange Memorandum of Understanding (MOU) between Fred Hutch and UWM went into effect on April 18, 2023, and called for an ongoing joint governance structure that oversees our data sharing collaboration relating to all Exchanged Clinical Data for Cancer-Related Care, Healthcare Operations, and Research. This governance structure works to define, in an ongoing way, the data that must be available for the adult oncology program.