艾昆纬-人工智能领域有前景的新发展-利用生成性和代理性人工智能进行创新，以提高临床试验的效率和质量（英）

White PaperPromising New Developments in Artificial Intelligence: Innovating with Generative and agentic AI to drive efficiency and quality in clinical trialsRAJNEESH PATIL, VP, Digital Innovation and Strategy, IQVIAPart 1 of a 4-part series highlighting key innovations in clinical trialsTable of contentsOverview 1How the bar was raised: ensuring safe and efficient use 2Selectivity of the training data 3Taking a multi-pronged approach to safeguarding efficiency and quality 3Curating and ‘containerizing’ data 4Integrating “human-in-the-loop” 4Harmonization of response 5Objectivity and context of use 5Recognizing uncertainty and knowledge gaps 5Putting principles into practice: use cases of successful utilization 6Successful utilization of a scientific chatbot in a Phase III trial 6Data review: empowering central monitors and CRAs 7Talking to your analytics: an agentic AI platform for monitoring support 8Looking ahead: supporting a shared vision to improve patient lives 8About the author 9 iqvia.com | 1OverviewAs implied by the article title — Promising New Developments in Artificial Intelligence — these are extraordinary times for the clinical research arena. Each year, with ever-increasing anticipation, thought leaders and industry professionals speculate and debate where and how to best unlock AI’s potential to help to make clinical trials more efficient and inclusive — and ultimately move healthcare forward by improving site and patient experiences. Within the vast and complex clinical trial ecosystem, there has been no dearth of opportunities to harness these remarkable technologies — whether in design and planning, patient engagement, trial delivery, clinical monitoring, regulatory submissions, or well beyond into the commercial drug development space.Recently, however, the landscape has transformed, thanks in no small part to the actual, successful application of GenAI tools into clinical trials — most of which were considered to be within the realm of ‘blue sky ideation’ not long ago. And while these advancements are proving to enhance both efficiency and quality in actual use cases, the FDA and other regulatory agencies are beginning to provide guidance on proper guardrails, context and validation requirements alongside other key considerations to incorporate Responsible-AI into clinical studies.In this paper, we will review several use cases that leverage IQVIA Healthcare-grade AI® engineered to meet the level of precision, speed and trust required in life sciences and healthcare, provide an overview of the technological and operational drivers of their success, and specifically address how each of these new AI methodologies substantially enhanced efficiency and quality.“Recently, the landscape has transformed, thanks in no small part to the actual, successful application of GenAI tools into clinical trials — most of which were considered to be within the realm of ‘blue sky ideation’ not long ago.”2 | Promising New Developments in A