Dr. Florian König-Huber works at the intersection of robotics, data science, and human-relevant biology as Head of Digital Transformation, Automation and Data Science at TissUse GmbH in Berlin. After beginning his career in Industry 4.0 automation, he brought his experience into life sciences to help improve the reliability and interpretability of complex biological experiments. He led the development of the first automated Multi-Organ-Chip system and now focuses on creating standardized and continuously monitored MPS workflows. His goal is to enable high-quality data at scale and to integrate machine learning models that support mechanistic understanding and digital twins in toxicology. He is motivated by accelerating the shift toward animal-free and more predictive drug development.
Towards Digital Drug Development: Integrating Multi-Organ Chips, Automation, and AI for Standardized High-Content Toxicology Data
Preclinical drug development relies on animal studies and conventional cell culture models, although both exhibit limited human relevance, with translational failure rates exceeding 85%. Microphysiological Systems (MPS) and Multi-Organ-Chip platforms aim to narrow this gap by enabling human organoid-based, systemic modeling. However, the increased biological complexity of these systems introduces significant demands on experimental reproducibility, data quality, and continuous monitoring. We demonstrate how robotic automation, standardized workflows, and integrated data infrastructures enable reproducible long-term MPS experiments at scale. Beyond reducing manual labor, automation establishes the foundation for data standardization and systematic AI-driven analysis. By combining continuous sensor data, imaging, and biochemical readouts with hybrid machine learning models and expert-guided soft sensors, we outline a pathway toward digital twins of organ-chip systems for toxicity prediction. Looking ahead, we discuss how such automated and data-rich MPS environments may support more coordinated research efforts and shared learning, pointing toward increasingly connected and scalable human-relevant toxicology.