Impact of AI on Clinical Development

PSI Mentoring 2025

Date: Ongoing 6 month cycle beginning late April/early May 2024 Are you a member of PSI looking to further your career or help develop others - why not sign up to the PSI Mentoring scheme? You can expand your network, improve your leadership skills and learn from more senior colleagues in the industry.

PSI Training Course: Mixed Models and Repeated Measures

This course is presented through lectures and practical sessions using SAS code. It is suitable for statisticians working on clinical trials, who already have a good understanding of linear and generalised linear models.

Joint PSI/EFSPI Visualisation SIG 'Wonderful Wednesday' Webinars

Our monthly webinar explores examples of innovative data visualisations relevant to our day to day work. Each month a new dataset is provided from a clinical trial or other relevant example, and participants are invited to submit a graphic that communicates interesting and relevant characteristics of the data.

PSI Career Young Virtual Event (Q1 2025)

This networking event is aimed at statisticians that are new to the pharmaceutical industry who wish to meet colleagues from different companies and backgrounds.

PSI Estimand Online Training (4 Day Workshop)

This is an interactive online training workshop providing an in-depth review of the estimand framework as laid out by ICH E9(R1) addendum with inputs from estimand experts, case studies, quizzes and opportunity for discussions. You will develop an estimand in a therapeutic area of interest to your company. In an online break-out room, you will join a series of team discussions to implement the estimand framework in a case study, aligning estimands, design, conduct, analysis, (assumptions + sensitivity analyses) to the clinical objective and therapeutic setting.

PSI Career Young Virtual Event (Q2 2025)

This networking event is aimed at statisticians that are new to the pharmaceutical industry who wish to meet colleagues from different companies and backgrounds.

PSI Career Young Virtual Event (Q3 2025)

This networking event is aimed at statisticians that are new to the pharmaceutical industry who wish to meet colleagues from different companies and backgrounds.

PSI Career Young Virtual Event (Q4 2025)

This networking event is aimed at statisticians that are new to the pharmaceutical industry who wish to meet colleagues from different companies and backgrounds.

Time	Agenda
09:00 - 09:30	Registration and Coffee
09:30 - 9:45	Welcome Karim Malki, Head, Predictive Analytics, UCB
09:45 - 10:45	Understanding activity patterns through wearable devices and AI algorithms Chris Holmes, Professor of Biostatistics in Genomics in the Nuffield Department of Clinical Medicine and the Department of Statistics, University of Oxford
10:45 - 11:30	Why statisticians should be driving machine learning and AI projects Dr Moira Verbelen, Principal Statistician, UCB
11:30 - 11:45	Break
11:45 – 12:30	The need for data science in the new clinical development paradigm Francis Kendall, Senior Director, Biostatistics and Programming, Cytel
12:30 - 13:30	Lunch and Networking
13:30 - 14:15	Technology Showcase Dr Bhushan Bonde, Head of IT - NewMeds Innovation Development, UCB
14:15 - 15:00	Predicting Crohn's disease risk in asymptomatic relatives Dr Ken Hanscombe, Research Associate at King’s College London
15:00 - 15:15	Coffee Break
15:15 - 16:00	Roche's experience developing Advanced Analytics Communities both internally and externally Chris Harbron, Expert Statistical Scientist, Roche
16:00 – 16:45	Panel Discussion
16:45 – 17:00	Closing Remarks
17:00 – 18:00	Drinks Reception and Networking

UCB	Technology Showcase Abstract: This will be a technical demonstration highlighting some of the latest technology available for performing machine learning and other high intensity computational tasks, including a demonstration of Intel’s portable neural compute stick and a look at how new advances, such as quantum computing, will change the technological landscape.
Chris Holmes (University of Oxford)	Understanding activity patterns through wearable devices and AI algorithms Abstract: New measurement technologies such as wearable devices coupled with AI algorithms, that can learn from large scale streaming data, have the potential for improved evaluation and monitoring of treatment interventions. In this talk we review the prospect for AI to better characterise population activity variation through wearable tech including an analysis of accelerometer data from 100,000 participants in UK BioBank.
Frances Kendall (Cytel)	The need for Data Science in the New Clinical Development Paradigm Abstract: This talk will set the stage on why Data Science is needed to support a New Clinical Development paradigm and what are the drivers of change. It will then put forward an idea on what that Paradigm could look like with examples of work that demonstrate this direction.
Moira Verbelen (UCB)	Why statisticians should be driving machine learning and AI projects Abstract: Advancements in computer science have popularised the widespread use of machine learning and AI. Although methods were mainly developed by computer and data scientists, they are rooted in statistical science. Statisticians are ideally placed to guide the implementation of these ‘new’ approaches in pharma. In-depth understanding of statistical concepts and model fitting are essential skills required to avoid pitfalls such as poor algorithm design, overfitting and incorrect interpretation of results. These considerations and the ensuing value of statisticians’ involvement are even more important in clinical development, where datasets tend to be smaller than those typically used for AI.
Ken Hanscombe (King’s College London)	Predicting Crohn's disease risk in asymptomatic relatives Abstract: An application of elastic net and random forest classifiers to Crohn’s disease (CD) risk in asymptomatic first-degree relatives (FDRs) of CD patients, using multiple environmental and genetic predictors.
Chris Harbron (Roche)	Roche's experience developing Advanced Analytics Communities both internally and externally Abstract: Roche has been successful in building an internal advanced analytics community consisting of over 750 data scientists from across the global Roche organization as well as establishing a number of external advanced analytics partnerships. This talk will discuss how Roche have approached this effort as well highlighting some of the successes and challenges, including crowd sourcing the internal community to tackle key scientific research questions using machine learning.

Time	Agenda
09:00 - 09:30	Registration and Coffee
09:30 - 9:45	Welcome Karim Malki, Head, Predictive Analytics, UCB
09:45 - 10:45	Understanding activity patterns through wearable devices and AI algorithms Chris Holmes, Professor of Biostatistics in Genomics in the Nuffield Department of Clinical Medicine and the Department of Statistics, University of Oxford
10:45 - 11:30	Why statisticians should be driving machine learning and AI projects Dr Moira Verbelen, Principal Statistician, UCB
11:30 - 11:45	Break
11:45 – 12:30	The need for data science in the new clinical development paradigm Francis Kendall, Senior Director, Biostatistics and Programming, Cytel
12:30 - 13:30	Lunch and Networking
13:30 - 14:15	Technology Showcase Dr Bhushan Bonde, Head of IT - NewMeds Innovation Development, UCB
14:15 - 15:00	Predicting Crohn's disease risk in asymptomatic relatives Dr Ken Hanscombe, Research Associate at King’s College London
15:00 - 15:15	Coffee Break
15:15 - 16:00	Roche's experience developing Advanced Analytics Communities both internally and externally Chris Harbron, Expert Statistical Scientist, Roche
16:00 – 16:45	Panel Discussion
16:45 – 17:00	Closing Remarks
17:00 – 18:00	Drinks Reception and Networking

UCB	Technology Showcase Abstract: This will be a technical demonstration highlighting some of the latest technology available for performing machine learning and other high intensity computational tasks, including a demonstration of Intel’s portable neural compute stick and a look at how new advances, such as quantum computing, will change the technological landscape.
Chris Holmes (University of Oxford)	Understanding activity patterns through wearable devices and AI algorithms Abstract: New measurement technologies such as wearable devices coupled with AI algorithms, that can learn from large scale streaming data, have the potential for improved evaluation and monitoring of treatment interventions. In this talk we review the prospect for AI to better characterise population activity variation through wearable tech including an analysis of accelerometer data from 100,000 participants in UK BioBank.
Frances Kendall (Cytel)	The need for Data Science in the New Clinical Development Paradigm Abstract: This talk will set the stage on why Data Science is needed to support a New Clinical Development paradigm and what are the drivers of change. It will then put forward an idea on what that Paradigm could look like with examples of work that demonstrate this direction.
Moira Verbelen (UCB)	Why statisticians should be driving machine learning and AI projects Abstract: Advancements in computer science have popularised the widespread use of machine learning and AI. Although methods were mainly developed by computer and data scientists, they are rooted in statistical science. Statisticians are ideally placed to guide the implementation of these ‘new’ approaches in pharma. In-depth understanding of statistical concepts and model fitting are essential skills required to avoid pitfalls such as poor algorithm design, overfitting and incorrect interpretation of results. These considerations and the ensuing value of statisticians’ involvement are even more important in clinical development, where datasets tend to be smaller than those typically used for AI.
Ken Hanscombe (King’s College London)	Predicting Crohn's disease risk in asymptomatic relatives Abstract: An application of elastic net and random forest classifiers to Crohn’s disease (CD) risk in asymptomatic first-degree relatives (FDRs) of CD patients, using multiple environmental and genetic predictors.
Chris Harbron (Roche)	Roche's experience developing Advanced Analytics Communities both internally and externally Abstract: Roche has been successful in building an internal advanced analytics community consisting of over 750 data scientists from across the global Roche organization as well as establishing a number of external advanced analytics partnerships. This talk will discuss how Roche have approached this effort as well highlighting some of the successes and challenges, including crowd sourcing the internal community to tackle key scientific research questions using machine learning.

Time	Agenda
09:00 - 09:30	Registration and Coffee
09:30 - 9:45	Welcome Karim Malki, Head, Predictive Analytics, UCB
09:45 - 10:45	Understanding activity patterns through wearable devices and AI algorithms Chris Holmes, Professor of Biostatistics in Genomics in the Nuffield Department of Clinical Medicine and the Department of Statistics, University of Oxford
10:45 - 11:30	Why statisticians should be driving machine learning and AI projects Dr Moira Verbelen, Principal Statistician, UCB
11:30 - 11:45	Break
11:45 – 12:30	The need for data science in the new clinical development paradigm Francis Kendall, Senior Director, Biostatistics and Programming, Cytel
12:30 - 13:30	Lunch and Networking
13:30 - 14:15	Technology Showcase Dr Bhushan Bonde, Head of IT - NewMeds Innovation Development, UCB
14:15 - 15:00	Predicting Crohn's disease risk in asymptomatic relatives Dr Ken Hanscombe, Research Associate at King’s College London
15:00 - 15:15	Coffee Break
15:15 - 16:00	Roche's experience developing Advanced Analytics Communities both internally and externally Chris Harbron, Expert Statistical Scientist, Roche
16:00 – 16:45	Panel Discussion
16:45 – 17:00	Closing Remarks
17:00 – 18:00	Drinks Reception and Networking

Past PSI Events

On this page:

Conferences