AI for Health

APRIL 12th

Understanding the many challenges inside clinical trials’ value chain, unveiling the potential of AI and data management to adress them, and exploring how it paves the way for coming innovative solutions.


Introduction of the Masterclass

Stephanie Trang, AI for Health - Master of Ceremony of the Masterclass

AI for evidence-based covariate adjustment in clinical trials
Félix Balazard, Owkin - In clinical trials, a lot of variables and bias can affect the statistical power (i.e. the study success' probability) of the efficacy analysis of the treatment. Félix Balazard presented us how Owkin’s methodology uses evidence-based covariate adjustment in order to increase the statistical power without need of a larger sample of patients.
In a nutshell, the principle introduced allows to reduce the noise partly due to the variance of the sample by using prognostic covariates associated with the endpoint of the clinical trial considering various parameters such as the cumulative incidence. AI for evidence-based covariate adjustment also allows to have less restrictive eligibility criteria for taking part in clinical trials, which means faster and wider involvement as for better generalization of the trial’s results.
The use of AI for evidence-based covariate adjustment provided by Owkin allows to deepen the understanding and the exploitation of data that could have been ignored or undetected. This shows that various problems and limits can be addressed by using modern AI technologies that can easily match and complete other evolutions in the clinical trials sector.

How to identify the right patients from a data perspective ?
Guillaume Carbonneau, Novartis - Targeting, profiling and localising clinical trials' participants is a challenge, especially in a world with an increase in targeted medicines and a need for population representativeness around the world.
Guillaume Carbonneau spoke about a methodology used in Novartis to target, profile and localise to improve patients matching to clinical research programs, improve investigators' experience and accelerate personalized medicine. Using big data is a challenge : having a lot of data is a good thing but harmonizing the data to be able to analyze it properly is the objective. Thanks to a cross functional team, they developed a cognitive intelligence-based platform linking AI and human intelligence which was presented to us through an use case about identifying and recruiting patients with a rare specific disease for an experimental treatment.
Big data in clinical trials needs harmonization and use efficiency for all the stakeholders in a clinical trial. Taking advantage of this to improve patients matching, investigator experience and personalized medicine is crucial in the present world's research landscape.

AI and Data Management for Clinical Trial Optimization
Linda Kallfa, NetApp - As clinical studies continue to become more complex it is important that the data generated is managed in the optimal way during the trial.
Linda Kallfa from NetApp explained to us the role of their technology in the drug development process and how implementing AI can lead the pharmaceutical treatment to the market faster. Clinical trials have different steps containing their own data sourcing from different activities with various storage and AI needs. In clinical trials, AI & ML are now widely used : to use data from patients (both structured and unstructured), create comparative models, use natural language processing, create digital twins in randomized controlled trials,...
Linda then identified three keys to AI success in clinical trials : the ability to manage a vast amount of data, a seamless data movement and the global speed of processing. Data management in its globality is then the challenge addressed by NetApp and the data fabric based on its technology.
NetApp brings modernization of their clients' data center, better exploitation of their cloud and acceleration in the distribution of their applications in order to improve data management in all the clinical trials' steps.

Clinical trials in 2030 : which perspectives for patients and pharmas with technological breakthroughs ?
Ariane Galaup, LEEM - How can we imagine the future clinical trials with all the lessons we have learnt due to the context and other major breakthroughs?
Ariane Galaup from Leem allows us to understand 6 major evolutions that will guide the innovations in clinical trials worldwide : ATMPs, biomarkers, real-world data, regional structure, decentralisation and patient-centric solutions.A focus has been made on decentralisation and patient-centric solutions, evolutions that were really increased because of the COVID-19 crisis. Finally, France has been described as an open country for innovation with real stakeholder commitment and would also take advantage of this to become the EU leader in innovation clinical trials.