2025 Global Summit

More information coming soon!

More information coming soon!

More information coming soon!

Sharing insights from the development of ML models to predict launch timelines using clinical trials data and regulatory approvals, with learnings from implementation of diverse approaches to improve data preparation and model outcome. The presentation will also highlight how the output can be used in launch or strategic forecasting to improve scenario design.

More information coming soon!

The use of Generative AI, specifically LLMs, can help pharmaceutical teams extract insights from large, structured datasets to support decision-making. Generative AI enables non-technical users to access data-driven insights with ease, improving market analysis, patient outcomes, and regulatory reporting. However, implementing LLMs for structured pharmaceutical data poses challenges, including complex data schemas, high query complexity, context limitations, latency requirements, and domain knowledge gaps. To address these challenges, schema-aware prompting techniques, iterative database exploration with agents, intermediate representations, and robust validation systems are employed. By sharing practical insights, this presentation aims to help pharmaceutical data professionals implement and scale Generative AI applications to enable agile, analytics-driven strategies, leading to faster, globally informed decision-making.

In this presentation, we will explore the innovative use of transformer models and image data to enhance patient event prediction in healthcare, with a focus on rare diseases. Traditional data sources like claims often fall short in identifying specific patient populations for personalized treatment strategies. Our study addresses these challenges by leveraging GenAI and transformer-based algorithms to process sequential data, offering new insights and improving productivity, revenue, and process transformation in healthcare.

We will discuss the development and fine-tuning of RareBERT, a rare disease-specific transformer model, using six years of data. By integrating image-extracted features from scanned EOB documents, RareBERT significantly improves the accuracy of predicting patient events, achieving a hit rate three times better than random predictions. Attendees will gain insights into the methodology, results, and the potential of transformer models in advancing healthcare research and patient care.

ETL is possibly the most time intensive part of analytics. Documenting the process, exceptions, business rules is never enjoyable, yet necessary. As data becomes increasingly dynamic, Extract, Transform, Load (ETL) processes require efficient solutions that maintain data quality, accessibility, and compliance. Generative AI (GenAI) is emerging as a transformative tool in ETL, enabling intelligent automation, standardized data handling, and streamlined documentation.

GenAI has the capability to apply uniform processes to non-standardized data sets- this is especially valuable. Organizations often work with data from diverse sources. For the sake of this workshop, we will use marketing data as a backdrop. Each vendor has a different data structure. Each department within the organization – finance, sales, CRM has a different data structure. Each time a new source is added- new processes must be created to integrate the data. So far, this is done by a team of engineers. GenAI can intelligently interpret and transform these disparate data sets, ensuring seamless integration across varied formats and minimizing manual adjustments.

Beyond this, GenAI automates the generation of metadata and can also automatically assign critical attributes—such as therapy area, brand lifecycle, and customer segment, marketing channel—based on the content and context of the data. This automated tagging further enhances usability across teams, and broadens the scope of analytics possible. Integrating GenAI into ETL processes creates a more agile and transparent data environment, where non-standard data is standardized, and critical attributes are automatically assigned. This abstract explores how GenAI can transform ETL workflows by simplifying data integration, enhancing traceability, and supporting compliance, ultimately fostering a more data-driven and efficient organization.

Learning Objectives

• Discover GenAI's role in automating ETL for non-standard data
• What kind of prompt engineering would you apply here?
• What are the limitations of using GenAI in ETL. It is not a magic wand.

This presentation explores how Generative AI is revolutionizing Real-World Evidence (RWE) by accelerating insights, optimizing clinical trials, and enabling personalized medicine. It highlights the role of cutting-edge technologies such as Agentic AI for autonomous hypothesis testing, DistilBERT for scalable NLP-driven text mining, and Multimodal AI for integrating diverse datasets like genomics, wearables, and clinical notes. The presentation also delves into Digital Twins, Federated Learning, Graph Neural Networks (GNNs), and Generative Adversarial Networks (GANs), showcasing their impact on synthetic data generation, privacy compliance, and predictive analytics. The session outlines challenges like privacy and regulatory compliance, offering solutions through Explainable AI (XAI) and federated learning. Attendees will leave with strategic recommendations for adopting AI-driven RWE platforms, upskilling teams, and driving healthcare transformation to deliver faster, more reliable evidence generation.

More information coming soon!

More information coming soon!

In today’s data-driven landscape, ensuring data quality (DQ) is essential for accurate decision-making and operational efficiency. However, traditional DQ methods are often reactive, addressing issues only after they arise, leading to delays and resource strain. To address this, we have developed a proactive Data Quality Framework that integrates structured interventions at every stage of data management to prevent issues before they impact operations. By leveraging Gen AI capabilities, this framework automates DQ checks and predictive analytics, creating a robust solution that strengthens data integrity and reliability across complex, multi-source processes.

More information coming soon!

More information coming soon!

This study incorporated the use of LLM to extract relevant information from the clinical notes. We explored a variety of prompts to generate a structured output that accurately captured the dates and corresponding medications for a patient’s treatment in oncology. The goal was to identify a prompt that would minimize hallucination and produce concise, structured responses. The optimal prompt that emerged from this exploration instructed the model to act as an assistant providing a therapy summary focused solely on relevant treatments. The prompt specifically guided the model to list the therapies by date or year in a numbered format, ensuring that only the information directly related to treatments of interest was extracted, without any additional or extraneous details. This approach effectively reduced hallucination and ensured that the output was more precise and relevant.

A Generative AI–enabled PMR solution that automates multi-format data processing (text, audio, video), uses Graph-based RAG with domain ontologies for high-accuracy insights, and accelerates decision-making across pharmaceutical therapeutic areas.

Segmentation and targeting processes are the real-world manifestations of pharma sales force strategy. Using reps’ time wisely to target the most valuable customers with precise messaging is critical to commercial success. Pharmaceutical companies use multiple data sources and advanced analytical techniques to ensure their reps get a competitive advantage when selling products.

Our presentation will address how sales reps can deeply improve their understanding of their customers and make them more productive in the field. It will also show how a harmonized framework for Generative AI-driven segmentation can ensure standards and address local market nuances.

More information coming soon!

This abstract presents a GenAI-powered automation framework that streamlines workflows in the pharmaceutical industry by automating tasks across platforms like Jira, Confluence, Git and Azure. Leveraging advanced AI techniques, it accelerates project timelines, enhances code quality, and improves decision-making, driving global productivity and efficiency.

In today's data-driven pharmaceutical industry, the ability to derive actionable insights from vast datasets is critical. Traditional dashboards often provide a static view, limiting the potential for dynamic and in-depth analysis. This limitation means that sales representatives, marketers, and leadership teams often lack the specific, real-time insights needed to make informed decisions.

Our presentation introduces the GenAI Copilot for Insights and Analytics, a sophisticated Generative AI tool designed to enhance how healthcare data—including EHR, prescriptions, clinical trials, and marketing activity data—is utilized. The GenAI Copilot leverages Large Language Models (LLMs) to process and synthesize data from multiple sources, providing real-time insights that are readily accessible. This tool empowers sales representatives and marketers by offering a better understanding of market trends and HCP activities, enabling more informed decision-making. The GenAI Copilot provides tailored insights that cater to the specific requirements of each role in Commercial Analytics teams, thus limiting the need for multiple static dashboards with a single, adaptive solution.

We will explore the current landscape in pharma, the limitations of existing tools, and how the GenAI Copilot addresses these challenges. Participants will gain an understanding of how this technology can drive improvements in commercial insights and analytics, fostering more effective strategies and outcomes.

More information coming soon!

Smart Assist is a conversational GEN AI assistant that enables your field teams to pull real-time insights on payouts, contests, and eligibility, streamlining performance tracking. It improves operational efficiency by making essential information accessible on the go, saving time and effort for field reps, and allowing them to focus on their sales goals. ZAIDYN™ Augmented Analytics is a Gen-AI driven tool to derive reporting insights, create custom KPIs, build reports, perform data analyses, generate data visualizations by using simple textual prompts. It also helps derive valuable patient journey related insights, highlight custom KPIs, build visualizations to aid decision making by using simple textual prompts enhancing patient-focused outcomes.