We will have two (parallel) short courses on August 30, 2022.
Short Course “Introduction to Supervised Statistical/Machine Learning Methods with Applications to Biomedical and Public Health Data”
Adam Ciarleglio, Department of Biostatistics and Bioinformatics, Milken Institute School of Public Health, The George Washington University, Washington, DC, USA.
Times: 8 AM – 12 PM and 1 PM – 5 PM
This short course will introduce fundamental concepts and methods of supervised statistical/machine learning with applications to biomedical and public health data. The course will focus on some of the most commonly utilized supervised learning methods for numerical and categorical responses including penalized regression methods, classification and regression trees, random forests, support vector machines, and neural networks (deep learning). There will also be a brief discussion of ensemble methods. Methods for model evaluation and assessment of variable importance will also be discussed as new algorithms are introduced. This course will employ a combination of lectures and hands-on R sessions during which participants will have the opportunity to put the lecture material into practice. This course is intended for those with some background in statistics (e.g., basics of statistical inference, linear and logistic regression) and a working knowledge of R who would like to be introduced to statistical/machine learning or would like to learn how to apply statistical/machine learning methods in their research. Participants should bring a personal laptop with R/RStudio installed prior to the start of the course.
About the lecturer:
Dr. Adam Ciarleglio is an Assistant Professor of Biostatistics in the Department of Biostatistics and Bioinformatics of the Milken Institute School of Public Health at The George Washington University in Washington, DC. He received his PhD in biostatistics from Columbia University in 2013 and completed a postdoctoral fellowship in the Division of Biostatistics in the Department of Child and Adolescent Psychiatry at the New York University School of Medicine in 2016. His current research focuses on bringing together the fields of machine learning and functional data analysis to develop tools for understanding heterogeneous effects of treatment using high-dimensional data. He is interested in using tools from these areas to help collaborators understand what characteristics (e.g., clinical, demographic, neuroimaging measures) influence the effects of antidepressant treatment in patients with major depressive disorder. He is also interested in methods for handling missing data in the context of functional data analysis. Dr. Ciarleglio has served as a statistician on projects in different areas of mental health research including depression, schizophrenia, and cognitive decline. He is also broadly interested in clinical trials, statistical computing, and statistics education.
Short Course “Platform Trials”
organized by: EU-PEARL (Franz König and Martin Posch, Medical University of Vienna)
Times: 9AM – 12.15PM and 1.30PM – 5PM
Topics of the course:
- General introduction to the problem:
Statistical Issues (multiplicity, sharing of data, etc)
Case studies from the past
- Introduction of specific case study for MCP conference
- Vanilla design
- Regulatory consideration on vanilla design
Topic 1: Error rates in platform trials
Regulatory on error rates
Topic 2: Increasing efficiency
Response adaptive randomization
Sharing of data
Regulatory considerations on these aspects
- Operational aspects