Flatiron Health Demonstrates Global Leadership in Responsible AI and Real-World Data Innovation at ESMO AI & Digital Oncology Congress 2025
Flatiron Health today announced its participation at the European Society for Medical Oncology (ESMO) AI & Digital Oncology Congress 2025, taking place November 12–14 in Berlin, Germany. Flatiron’s presence at the event highlights its deep methodological expertise in artificial intelligence (AI) and machine learning (ML) applications across global markets and underscores the company’s commitment to advancing trustworthy, large language model (LLM)-extracted real-world data (RWD) at scale.
At ESMO AI 2025, Flatiron experts will present five original research studies and lead a featured speaking session titled “The VALID Framework for LLM-Extracted Oncology Data Quality: Comprehensive Methodology for Evaluating LLM-Extracted Real-World Oncology Data Quality.” The company’s contributions to the conference reflect a growing emphasis on data reliability, validation methodologies, and responsible AI integration into clinical oncology workflows.
Our research presented at ESMO AI demonstrates how LLM-extracted real-world data can be rigorously validated across diverse healthcare systems and countries,” said Nathan Hubbard, Chief Executive Officer at Flatiron Health. “These efforts empower our partners to generate credible, fit-for-purpose evidence that informs and advances cancer care and research worldwide. Flatiron’s team continues to define what responsible AI in oncology looks like—advancing the field through methodological rigor, privacy-preserving innovation, and frameworks that make AI trustworthy and scalable.”
Pioneering AI-Driven Real-World Data Quality
Flatiron Health’s participation at ESMO AI 2025 comes at a pivotal moment in the evolution of oncology research, as the industry increasingly turns to AI-driven real-world evidence (RWE) to complement traditional clinical data. Flatiron’s work focuses on developing and validating LLM-based extraction technologies that transform unstructured clinical notes into structured, analyzable data without compromising accuracy, transparency, or patient privacy.
Among the company’s featured research, Flatiron is presenting its first published study on digital twin modeling—an innovative approach that leverages LLM-extracted clinical context to enable more personalized survival predictions and treatment planning. By incorporating real-world variables and AI-powered insights, these models aim to help clinicians and researchers simulate individualized treatment outcomes and optimize therapeutic decisions.
In addition, multiple studies will showcase Flatiron’s advances in data validation and quality assessment for AI-extracted oncology data across tumor types, healthcare systems, and geographies. The company’s commitment to reproducibility and transparency ensures that AI-derived insights maintain the same standards of rigor traditionally reserved for manually curated datasets.
The VALID Framework: Setting a New Standard for AI Data Evaluation
A central focus of Flatiron’s presence at ESMO AI 2025 is its VALID Framework—short for Validation of Accuracy for LLM/ML-Extracted Information and Data. This framework represents the industry’s first comprehensive, systematic approach to evaluating AI-extracted real-world data quality.
The VALID Framework outlines principles and metrics for assessing accuracy, completeness, and reproducibility in AI-extracted oncology datasets. It enables researchers, regulators, and healthcare organizations to confidently measure the reliability of data derived from AI models and integrate these insights into evidence generation, clinical decision-making, and regulatory submissions.
Flatiron’s leadership session at ESMO AI will explore the methodological underpinnings of the VALID Framework, sharing case studies from real-world applications that demonstrate how the framework can be deployed to evaluate model performance across populations, tumor types, and healthcare systems. The session will also highlight how these practices align with Flatiron’s broader vision of advancing a responsible, transparent, and trustworthy AI ecosystem in oncology.
Driving Global Collaboration and Innovation
Flatiron’s engagement at ESMO AI 2025 reinforces its role as a global partner to researchers, biopharmaceutical companies, and healthcare providers who seek to leverage real-world data for more efficient and inclusive cancer research. By combining advanced analytics with deep domain expertise, the company continues to lead the industry toward an era of AI-enhanced evidence generation that transcends traditional data boundaries.
Through its collaborations across North America, Europe, and Asia, Flatiron is actively working to establish consistent data standards and validation protocols that facilitate global interoperability and enhance the credibility of AI-generated insights. These initiatives are essential for enabling real-world data to play a transformative role in clinical research, regulatory decision-making, and value-based healthcare.
Our work at ESMO AI 2025 reflects Flatiron’s unwavering commitment to building the frameworks, methodologies, and partnerships that define responsible AI in medicine,” added Hubbard. “By focusing on trust, transparency, and scalability, we are shaping a future where AI not only accelerates discovery but also enhances patient outcomes worldwide.”
About Flatiron
Flatiron Health is a healthtech company expanding the possibilities for point of care solutions in oncology and using data for good to power smarter care for every person with cancer. Through machine learning and AI, real-world evidence, and breakthroughs in clinical trials, we continue to transform patients’ real-life experiences into knowledge and create a more modern, connected oncology ecosystem. Flatiron Health is an independent affiliate of the Roche Group.
source link: https://www.businesswire.com
