REGISTER NOW! https://www.go-acop.org/

• 04 Nov 2023 (8:15 – 17:00 ET)
• 05 Nov 2023 (8:15 – 17:00 ET)

This hands-on workshop will take us through the core concepts behind DeepPumas and how this tool can bridge the gap between mechanistic, statistical and machine learning modelling. We will start by running through some basic pharmacometric models and workflows within the Pumas cloud-based software suite to get everyone familiar with the tools we use. We will then successively build up our understanding of machine learning models and how they can be integrated in various ways into what’s now semi-mechanistic pharmacometric models. We will explore the current state of machine learning, the basics of machine learning based regression for function identification, embedding these regressors into dynamical systems to generate neural differential equations and universal differential equations, using mixed effects with neural networks to enable identification of individualizable functions, and tying it up with machine learning based identification of prognostic factors from covariate data.

Taken together, this technology of seamless embedding of machine learning methodology into nonlinear mixed effect models is immensely powerful for model identification and for simultaneous utilization of our domain-specific knowledge and of the data we might have available. After this workshop, participants will have a grasp of what can be done, how this might affect the healthcare space going forward, and hands on experience in doing hybrid mechanistic and machine learning modeling themselves.

Faculty:

• Niklas Korsbo, PhD
• Mohamed Tarek, PhD
• Jose Storopoli, PhD

Learning Objectives:

1. Learn basic usage of the Pumas and DeepPumas software.
2. Understand simple neural networks and how to use them.
3. Understand neural-embedded dynamical systems such as Neural Differential Equations (ODEs) and Universal Differential Equations (UDEs).
4. Learn Mixed-Effect Neural Networks (MeNets) and how to find individualizable models.
5. Learn how to integrate machine learning for prognostic factor identification in longitudinal models.
6. Understand how to compose all the above to enable a semi-mechanistic scientific machine learning approach to finding individualizable and predictive longitudinal models.
7. Get a sense of how this technology may transform the role of data and modeling in the healthcare space.

REGISTER NOW! https://www.go-acop.org/

Agenda:

04 Nov 2023 (8:15 – 17:00 ET)

05 Nov 2023 (8:15 – 17:00 ET)

REGISTER NOW! https://www.go-acop.org/