As scientific discovery expands and regulatory timelines tighten, life sciences teams are under growing pressure to generate high-quality evidence faster and at scale. But even as new tools emerge to support drug development, persistent challenges remain — from manual systematic literature reviews (SLRs) and complex trial cohorting to inconsistent real-world data (RWD). These inefficiencies not only slow research; they limit reproducibility and increase costs.
To close the gap between discovery and delivery, organizations are turning to AI-driven solutions grounded in standardized clinical terminology. Together, these technologies create a scalable foundation for evidence generation — one that accelerates literature review, streamlines patient identification, and brings trustworthy insights to market sooner.
In our latest insight brief, Life sciences and the need for speed: How AI and clinical terminology accelerate evidence generation, we explore how this pairing transforms SLRs, clinical trial cohorting, and RWD analysis to improve precision and reproducibility at every stage of research.
INSIGHT BRIEF
Life sciences and the need for speed: How AI and clinical terminology accelerate evidence generation
Below is an excerpt from one of the recent studies featured in the brief.
Advancing SLR efficiency with generative AI and clinical terminology
Regeneron Pharmaceuticals faced a bottleneck that similar organizations often encounter — manual screening and data extraction that consumed months of expert time and limited the pace of new insights. To tackle this, IMO Health partnered with them to build on the success of the oncology collaboration and develop a next-generation, generative AI-based SLR solution. While the integration of clinical terminology is an active area of ongoing work, this study focused on evaluating the performance and reliability of the generative AI system itself. The model automated abstract and full-text screening, structured extracted data, and evaluated outputs with greater consistency and reproducibility.
Results and impact
The study, published in JAMIA in 2025, demonstrated that generative AI can deliver accuracy and reliability comparable to expert reviewers while dramatically reducing manual workload. Across multiple therapeutic areas, the system consistently performed well in abstract screening and data extraction, showing strong alignment with human reviewers’ decisions. The findings confirm that generative AI has the potential to significantly streamline the SLR process while maintaining scientific rigor and reproducibility.
Enhancing systematic literature reviews with generative artificial
intelligence: development, applications, and performance evaluation
JAMIA | April 2025
IMO Health authors Surabhi Datta, Majid Rastegar-Mojarad, Kyeryoung Lee,
Julie Glasgow, Chris Liston, Long He, Xiaoyan Wang
