Abstract: Traditional clinical trial monitoring, reliant on labor-intensive site visits and manual data review via Electronic Data Capture systems, is both time-consuming and resource-intensive. The emergence of risk-based monitoring (RBM) and quality tolerance limit (QTL) frameworks offers a more efficient alternative by proactively identifying systematic risks to patient safety and data integrity. In this paper, we propose an advanced machine learning (ML) approach to enable real-time, automated QTL risk assessment. The QTL-ML framework integrates multi-domain clinical data to predict diverse QTLs at the program, study, site, and patient levels, bypassing limitations of static thresholds and single-source data dependency. Our assumption-free approach leverages dynamically accumulating trial data to detect and mitigate risks in an automated manner. Embedded within ICH-E6(R2) RBM principles, this innovative solution enhances patient safety, reduces trial durations, and curbs costs. Furthermore, we introduce an extension leveraging a deep learning framework, incorporating hierarchical anomaly detection and temporal analysis to enhance accuracy and scalability across clinical trial settings. Together, these methodologies hold transformative potential for the efficacy and sustainability of clinical trial monitoring.

Abstract: Writing statistical analysis plans (SAPs) is often time-consuming, particularly for junior statisticians new to the pharmaceutical industry. As a standard internal practice, a well-developed SAP draft must be finalized before first subject in. Delays in this process can hinder the timely integration of statistical strategy into study design and execution. Furthermore, despite having a standard SAP template, inconsistencies in practice still occur—potentially compromising document quality and creating misinterpretations during collaboration with programming teams. With increasing demands to accelerate development while reducing cost and timelines, a smarter, faster, and more consistent approach is urgently needed. We present a retrieval-augmented generation (RAG) powered large language model (LLM) system that automatically generates a protocol-specific SAP. To address the challenge of LLM hallucinations, we explored multiple optimization strategies within the RAG architecture to ensure stable and reliable output. Our approach leverages semantic search, vector embeddings, and domain-tuned retrieval to extract protocol-specific details, which are then integrated into a pre-aligned SAP template through the LLM. The resulting system delivers measurable improvements in efficiency, cost savings, and compliance—while maintaining scientific rigor through human-in-the-loop interactive review.

Abstract: Artificial intelligence (AI) is transforming the pharmaceutical industry across the entire drug development lifecycle—from target identification and preclinical research to clinical trials, pharmacovigilance, and advanced manufacturing. This presentation provides an overview of the evolving AI landscape, highlighting investment trends, regulatory perspectives, and real-world applications. Key advances include deep learning for drug discovery, natural language processing for pharmacovigilance, and machine learning methods such as similarity-based approaches that address classical statistical paradoxes in clinical trial design. Regulatory initiatives by the FDA and EMA are shaping the adoption of AI tools, particularly in digital twins and precision medicine. Challenges remain, including data quality, privacy, interpretability, and resistance to paradigm shifts away from traditional statistics. Through case studies such as similarity-based machine learning, Procova models, and AI-driven digital twins, this talk illustrates how AI can reduce errors, improve predictive accuracy, and enable personalized medicine. The discussion concludes with broader implications of AI in healthcare and education, emphasizing critical thinking and creative analogies as essential skills for navigating the AI era.

Abstract: Large language models (LLMs) have shown impressive fluency in biomedical text, but their limitations in factual accuracy and scientific reasoning remain major barriers for drug discovery. We present IKraph, a large-scale, high-quality biomedical knowledge graph built with our methods that won the LitCoin NLP and BioCreative Challenges. Equipped with a highly explainable method, Probabilistic Semantic Reasoning (PSR), IKraph enables robust reasoning for drug development applications such as drug repurposing. On a benchmark evaluation, IKraph significantly outperformed leading LLMs and LLM-based systems including GPT-4o, Claude, Gemini and FutureHouse underscoring the advantages of knowledge graphs in delivering reliable and explainable insights. Building on IKraph, we are developing two new applications. IKnow (Integrated Knowledge Intelligence) is a fact-checking system that validates information from PubMed articles and newly submitted manuscripts against the entire body of knowledge captured from PubMed. IDEAL (Insilicom Data Exploration And Learning) integrates IKraph with harmonized genomics data to deliver accurate, data-driven answers to biomedical questions, powered by our BioASQ 13B award-winning biomedical QA method. These efforts highlight how knowledge graphs address key bottlenecks of current LLMs, offering a scalable and trustworthy foundation for scientific discovery. Our work demonstrates the critical role of knowledge graph–driven AI in accelerating drug development and advancing biomedical research.