AI and innovation in clinical trials
Aarav Badani, Fábio Ynoe de Moraes, Philipp Kickingereder, Caroline Chung, Alireza Mansouri
Abstract
Clinical trials face persistent challenges in cost, enrollment, and generalizability. This perspective examines how artificial intelligence (AI), large language models (LLMs), adaptive trial designs, and digital twins (DTs) can modernize trial design and execution. We detail AI-driven eligibility optimization, reinforcement learning for real-time adaptation, and in silico DT modeling. Methodological, regulatory, and ethical hurdles are addressed, emphasizing the need for validated, scalable frameworks to enable responsible and widespread integration.
Topics & Concepts
Clinical trialPerspective (graphical)Artificial intelligenceMedicineComputer scienceReinforcement learningFace (sociological concept)MEDLINEEngineering ethicsPsychologyData scienceMedical physicsResearch designMachine learningManagement scienceResearch ethicsRisk analysis (engineering)ScalabilityMultidisciplinary approachArtificial Intelligence in Healthcare and EducationMachine Learning in HealthcareAdvanced Causal Inference Techniques