AI Developers Can Now Build at Scale and Meet Emerging AI Assurance Standards, Putting Their Technology to Use to Improve Patient Outcomes

NEW YORK: Atropos Health, a leader in translating clinical data into high-quality, personalized, real-world evidence (RWE) for care, today announced the availability of AI model training on the Atropos Evidence Network, the largest federated healthcare data network with more than 300 million patient records. Atropos Evidence Network members can utilize certain de-identified real-world data (RWD) to train AI models and seamlessly deploy their technology via channel partner systems, including pharmaceutical companies and health systems like Stanford.

AI model training benefits multiple users across the healthcare ecosystem:

• AI Developers: can test their tools and applications (AI, ML, LLM, or NLP) on high-quality RWD available in the Atropos Evidence Network.
• Health Systems: can test in-house developed AI models on the largest federated data network in healthcare and become a channel partner of advanced models built by others, all while following the emerging assurance standards on AI transparency and bias detection.
• Life Sciences Organizations: can build, test, train, and distribute AI models to identify overlooked patient populations and leverage the Atropos Health network of providers to utilize the models.
• Partners and Data Holders: have a new use case for their data that can enhance and extend the value of their data and provide opportunities to improve their Real-World Data Score™(RWDS), driving monetization and value back to data sources without losing data possession and control.

Training AI models on RWD is imperative to the safety and efficacy of AI use in healthcare as companies continue to innovate at hyperspeed. Atropos Health is committed to the use of responsible AI and prioritizing patient safety, which is the driving principle in every solution. By training AI models on real-world data, healthcare organizations integrating AI tools can ensure their models output the most accurate recommendations to researchers and clinicians, ultimately improving patient care and outcomes.

“Our core GENEVA OS™ technology has now been deployed at dozens of healthcare and life science institutions as a standard in quality and automation for evidence and insight generation,” said Dr. Brigham Hyde, CEO and Co-Founder of Atropos Health. “Over the last several months we have extended our core vector database capability to enable classifier and AI model development, built on the same tenants of quality, transparency, and automation. This development enables our community of users and partners to leverage this core technology to develop, test, and train high-quality predictive models, removing the burdens of data acquisition, preparation, and infrastructure.”

Members utilizing Atropos Health's federated data network include a top-five pharmaceutical company and QuantHealth, an AI-powered drug development platform. QuantHealth is leveraging real-world data to enhance its Clinical-Simulator, which forecasts individual patient responses within clinical trials, enabling design teams to anticipate overall trial dynamics and adjust strategies accordingly.

“De-risking and optimizing clinical trials through robust patient-level simulations is no easy feat, which is why we have continued to evolve and mature our AI platform and underlying data frameworks,” said Orr Inbar, CEO and Co-Founder of QuantHealth. “By doing this, we’ve been able to help 7 of the top 20 pharma companies simulate and optimize their trials and clinical programs to ensure clinical and operational excellence. Our partnership with Atropos Health will allow us to conduct real-time simulations and deploy our models to the point of care, unlocking new opportunities and use cases for our pharma partners.”

Members of the Atropos Evidence Network gain immediate access to vector databases and Clinical Definitions Library (CDL) via Atropos Health GENEVA OS™. Ready-built for AI classifier development, the vector database and patient timelines are ideal for training classifiers. All data on the Atropos Evidence Network is represented as a Patient Object Vector and all data sets are organized under the same object-oriented schema.

This new capability follows Atropos Health’s Data Quality ScoreCard, which became available to members of the Atropos Evidence Network this fall. These scorecards provide data contributors with confidential, transparent, analytically driven feedback on their data quality, including where they rank compared to Atropos Evidence Network averages and suggestions for improving data quality.

About Atropos Health:

Atropos Health is the developer of GENEVA OS™, the operating system for rapid healthcare evidence across a robust network of real-world data. Healthcare and life science organizations work with Atropos Health to close evidence gaps from bench to bedside, improving individual patient outcomes with data-driven care, expediting research that advances the field of medicine, and more. We aim to transform healthcare with timely, relevant real-world evidence. To learn more about Atropos Health, visit www.atroposhealth.com.

Fonte: Business Wire