Durch KI Einsatz in der Wirkstoffforschung und klinischen Studien,
Produktivität steigern, 'time to market' beschleunigen und Kosten senken.
September 01 - 03, 2025 | Palace Hotel Berlin, Germany
2. Conference Day, 03. September 2025
9:10 - 9:20
Opening by Chair Jessica Cordes, Interim Head of Clinical Operations, Clinical Excellence
9:20 - 9:50
GenAI Meets RWD: Building a Multi-Agent GenAI Platform for RWD/RWE & Insight
Abhishek Choudhary -
Principal Data Engineer,
Bayer
-Bayer’s modular, multi-agent GenAI platform for extracting insights from Real-World Data (RWD) and Real-World Evidence (RWE).
-How specialized agents handle data curation, harmonization, and evidence generation across complex healthcare datasets.
-Governance, scalability, and compliance considerations for deploying GenAI in regulated pharma environments.
9:50 - 10:20
Before GenAI, Get Your Data Right: Why FAIR Data Matters
Selena Baset -
Senior Information Architect,
Roche
- The GenAI Promise: How AI is expected to transform clinical trials—faster decision-making, better predictions, and improved efficiency.
- The Risk of Bad Data: Without structured, high-quality, and governed data, Gen AI models risk hallucinations, bias, and unreliable outputs.
- FAIR Data as a Prerequisite: Findable, Accessible, Interoperable, and Reusable (FAIR) data principles lay the foundation for effective and trustworthy Gen AI. FAIR and Gen AI are not competing forces; they enhance each other. FAIR data enables AI to generate meaningful, actionable insights.
- The Roadmap Forward: Steps pharma organizations can take to ensure data governance and AI readiness for clinical trials.
10:20 - 10:50
Coffee Break & Networking
10:50 - 11:20
Panel: Data foundations for the continuous development and application of AI
Abhishek Choudhary -
Principal Data Engineer,
Bayer
Selena Baset - Senior Information Architect, Roche
Dimitri Metzger - Digital, Data & IT Business Partner for Development & Medical Units & AI, Merck
Jessica Cordes - INTERIM HEAD OF CLINICAL OPERATIONS, Clinical Excellence GmbH
-Data quality and integrity: Establish rigorous processes for data collection, review, and cleansing. High-quality data is essential for training reliable AI models.
-Data integration and management: Working with various data types, including chemical data, biological data, clinical outcomes, patient records, and realworld evidence. Integrate these data sources and establish effective data management systems and practices.
-Human-machine interaction: Preparing your organization for the integrated use of AI through data and AI competency programs, continuing education, and change management.
-Legal and ethical considerations: Understanding AI law. Ethical implications of AI, including bias in AI models and the impact of AI decisions in clinical settings.
Dimitri Metzger
Digital, Data & IT Business Partner for Development & Medical Units & AIMerck
11:20 - 11:50
In-House Generative AI Implementation: Practical Insights from a Mid-Sized Pharma Company
Dr. Joachim Hagel -
Medical AI Engineer,
Infectopharm Arzneimittel Und Consilium GmbH
- How to implement generative AI effectively and sustainably to improve efficiency and quality.
- Practical insights from an internal AI initiative – from pilot projects to strategic decisions and productive use.
- Key success factors: agile implementation, interdisciplinary collaboration, and structured change management.
- Opportunities, pitfalls, and proven solutions – what makes generative AI truly successful in practice.
11:50 - 12:20
AI in the Service of Quality
Dr. Colin Lischik -
Head of GQDT Lab Automate,
Boehringer Ingelheim Corporate Center GmbH
- Why use AI for quality management?
- Regulatory thresholds
- Bridging the gap between technology and domain experts
- Quality of the future
12:20 - 13:50
Lunch Break and Networking
13:50 - 14:20
Target discovery using GenAI, knowledge graphs and literature
Nikola Milosevic -
Science Fellow,
Bayer AG
-We have developed systems based on GenAI to rank and explain targets
-The system is based on literature, clinical trial documents, and knowledge graph
-Current state-of-the-art LLMs are able to analyze multi-modal data and provide comprehensive evaluation of the targets
-Pipeline can produce in few hours comprehensive summary of possible targets for a disease
14:20 - 14:50
AI in Proteomics: Transforming Clinical Research through Multi-Omics Data Integration• Recent advanced in biomarker discovery using multi-omics data and AI
Dr. Ornella Cominetti -
Senior Omics Data Specialist,
Nestlé
-Important role of biobanks and recent large proteomic studies/cohorts
-AI versus standard statistical models in the context of omics data analysis
-Examples in metabolic health, cancer, brain health, and infectious diseases
14:50 - 15:20
Machine Learning in Early Drug Discovery – How Bayer Uses Deep Learning to Explore Drug Candidates Simultaneously in Phenotypic, Chemical, and Transcriptional Spaces
Dr. Matthias Orlowski -
principal machine learning engineer,
Bayer Vital GmbH
- How Bayer Creates Large-Scale Cell-Painting Assays
- How to Use Machine Learning to Embed Microscopy Images for Phenotypic Profiling
- Machine Learning Models for Embedding Chemical Structures
- Multimodal Data Analysis
