The Collaborative Research Institute Intelligent Oncology (CRIION) combines experimental oncology, machine‑learning engineering, and a wet‑lab facility to build trustworthy, explainable AI models that integrate genomics, imaging, and clinical data. These models enable early cancer detection, precise patient stratification, and real‑time adaptive therapy design, supporting academic researchers, clinical departments, and biotech companies in biomarker discovery and treatment prediction.
Funding
Funding not disclosed
Founders
Product
Problem
Current cancer research and clinical practice often operate in silos, limiting the ability to integrate high‑dimensional biological data, advanced imaging, and real‑time patient information into actionable treatment decisions. This fragmentation hampers early detection, precise patient stratification, and the development of adaptive therapies.
Solution
The Collaborative Research Institute Intelligent Oncology (CRIION) creates a unified research environment that combines experimental oncology, machine‑learning engineering, and philosophical inquiry to develop trustworthy AI systems for cancer care. By operating both an AI lab and a wet‑lab facility in Freiburg, CRIION generates and validates data pipelines that translate raw molecular, imaging, and clinical datasets into predictive models. These models are designed to be robust, efficient, and explainable, enabling clinicians and researchers to identify hidden cellular states, predict treatment response, and continuously adapt therapeutic strategies as new data become available. The institute also supports collaborative projects, co‑financing, and joint development to accelerate the translation of AI‑driven insights into clinical practice.
Target Audience
Primary users are academic oncology research groups, clinical oncology departments, and pharmaceutical or biotech companies seeking AI‑enhanced biomarker discovery, treatment prediction, and adaptive therapy design.
Features
- Integrated AI and wet‑lab platforms that allow end‑to‑end development from data acquisition to model deployment
- Explainable‑AI frameworks (e.g., “One For All”) that jointly optimize classification and self‑explanation, increasing clinician trust
- Multi‑modal data handling for genomics, liquid biopsy, imaging, and unstructured clinical records
- Real‑time adaptive modeling pipelines that update predictions as patient data evolve
- Open collaborative model supporting co‑financing, joint project funding, and interdisciplinary partnerships
- Dedicated expertise in robust, efficient, and trustworthy machine‑learning methods tailored to oncology