Kidney Week 2025 Early Program - Advances in Research Conference 2025: Generative Artificial Intelligence in Kidney Disease Research and Management-Panel Discussion: Future of Clinical Trials

Video Transcription

Video Summary

The transcript captures a panel discussion on applying AI in clinical trials and healthcare workflows, especially beyond Epic-based systems. Speakers explain that interoperability mandates (HITECH/Cures Act) and standards like FHIR/HL7 enable tools to work across EHRs such as Cerner, eClinicalWorks, Athena, and Allscripts, though institutions still require security/IT access processes.<br /><br />Panelists discuss bias mitigation, arguing that “correcting” real-world underrepresentation is often better handled through intentional recruitment strategies and AI-enabled large-scale screening to prioritize outreach to underserved groups. On AI event adjudication, they suggest AI adds most value when adjudication is time-consuming and large-scale, favoring human-in-the-loop approaches, while fully automated adjudication might be acceptable only with very high accuracy and likely some auditing.<br /><br />Looking 10 years ahead, they predict trials will become cheaper, faster, and more “continuous,” using AI for protocol design, data-driven site feasibility, automated recruitment, broader access (translation, consent chatbots), remote monitoring, and faster evidence synthesis. Cautions include clinician distrust, confirmation bias from pre-filled summaries, LLM hallucinations, poor note quality/missingness, and the need for drift monitoring, retraining governance, and regulatory lag.

Asset Subtitle

Alexander J. Blood, Lili Chan, Samarra Badrouchi, Bashar Kadhim, Navdeep Tangri

Meta Tag

Module ARC

Speaker FACULTY FACULTY

Keywords

AI in clinical trials

Healthcare workflow automation

EHR interoperability

FHIR HL7 standards

Bias mitigation and recruitment

Human-in-the-loop adjudication

LLM governance and monitoring