In this session, we will describe Pfizer's AI journey through the lens of clinical data, use cases, implementation and key to success. Clinical Data Management for the Vaccine Study presented an opportunity for ML/NLP to assist in saving valuable time reconciling data. The foundation for a Smart Data Quality strategy was expanded to other TAs thanks to the solution's Pattern Recognition, Clinical Inference capabilities that will be explained in detail.

Once life sciences companies have proven the value and reliability of AI models, they need to deploy that insight to the right person at the right time to drive the right decision. This presentation will discuss how to implement AI in the workflow and discuss three examples where organizations have successfully done this.

Gaining insights from data has traditionally been a laborious and time-consuming effort. It includes ingestion of data from many sources, aggregation via programming, cleaning through listings review and validation checks, and provisioning of data to downstream stakeholders in various formats. This session explores the challenges with these processes and provides methods for automation with the use of artificial intelligence to accelerate access to downstream data consumers for quicker critical decision-making.

AI/ML is over-hyped, this panel will discuss machine learning techniques that are in production in various organizations that are adding value and accelerating Clinical Development. We will also discuss best practices, lessons learnt, how to pick a ML use case from idea to implementation and more.

Natural language understanding and knowledge graphs in pharma. Biomedical text mining is hard. Recent techniques, like transformers, trained on publically available data, like Pubmed, can give better language models for use in pharma. How do new techniques like transformers help with better language models? Knowledge graphs and graph convolutional network applications in pharma.

Over 80% of healthcare information is buried in unstructured data like provider notes, pathology results and genomics reports. Leveraging AI and NLP technologies to mine, contextualize and temporalize medical concepts can have a dramatic effect on clinical trial operations. We discuss how effective use of this information can accelerate multiple operational objectives across the clinical trial continuum such as study design, site selection, patient recruitment, SAE adjudication, RWE and beyond.

This presentation will discuss approaches and case studies for extracting knowledge from clinical trial data and connecting it with preclinical and post-approval data.

In feasibility, trial-sites are chosen based on medical expertise and patient access. Using principles of fairness in machine learning, a model that maps clinical trial descriptions to a ranked list of sites was developed and tested on real-world data. It resulted in a list of potential trial-sites that accounted for performance and diversity. This presentation looks at data sources and ML algorithms that could solve diversity problems in site selection.

The widespread adoption of electronic health records (EHRs) alongside the advent of scalable clinical molecular profiling technologies has created enormous opportunities for deepening our understanding of health and disease. However, in most diseases, disease-relevant markers are spread across multiple biological contexts that are observed independently with different measurement technologies and at various time schedules, and their manual interpretation is therefore in many cases complex. Machine learning holds promise for integrating comprehensive, deep phenotypic patient profiles across time for (i) predicting outcomes, (ii) identifying patient subtypes and (iii) associated biomarkers. In this talk, we will outline opportunities and challenges for clinical prediction models built from deep phenotypic patient profiles in clinical research and beyond.

Learn which AI-based technologies are in production for which ICSR process steps. Understand various considerations for planning, implementation, and validation. Understand key learnings from early adopters of AI-based technologies within the ICSR process.