Samuel Branders, Ph.D., Data Mining and Statistical Research Scientist of Tools4Patient (T4P) recently presented cutting-edge research at the 2018 Promoting Statistical Insight conference in Amsterdam. This annual meeting is centered around the current topics relating to best practices and current challenges in the pharmaceutical industry. Dr. Branders’ presentation was entitled “Leveraging Historical Data for High-Dimensional Regression Adjustment, a Machine Learning Approach” which described the T4P team’s approach of using mathematical modeling to enable clinical trials.
Data Driven Approach
This approach was originally developed based on the observation that the amount of data collected from patients involved in clinical trials is continuously growing, providing potential covariates that could be used to improve study analysis and power. The proposed approach utilizes historical data and modeling to build a composite covariate that is intended to improve statistical power and reduce the probability of a type II error in phase 2 and phase 3 trials. Tools4Patient is utilizing and further developing similar machine learning technologies in Placebell©TM, its novel approach for managing the placebo response in clinical trials. Placebell©TM combines a validated personality questionnaire and patient demographics with a proprietary algorithm to generate a composite Placebo-covariate, which can then be used as a baseline covariate in statistical analyses.
About Samuel Branders, Ph.D.
Dr. Branders is a graduate of the University of Louvain with a Ph.D. in Machine Learning and Bioinformatics. During his Ph.D., he worked on survival analysis applied to cancer research with both biomedical and methodological / statistical contributions. Dr. Branders’ main areas of interest are to optimize biological signal interpretation and by using predictive modeling contribute to better clinical outcome analysis. He is an integral part of the Tools4Patient internal research and development team, which is comprised of neurophysiologists, clinical scientists, mathematicians and statisticians
