Making Data Science Work for Clinical Reporting

Overview

Class Central Tips

This course is aimed to demonstate how principles and methods from data science can be applied in clinical reporting. By the end of the course, learners will understand what requirements there are in reporting clinical trials, and how they impact on how data science is used. The learner will see how they can work efficiently and effectively while still ensuring that they meet the needed standards.

Syllabus

Making Data Science work for clinical reporting

In this module we will introduce this course. We will provide context on clinical reporting in general, describing how clinical trials work at a high level, as well as providing resources to learn more. We will then focus on motivating the course, describing the benefits of applying data science in the context of clinical reporting

The burden of being faultless and transparent

In this module we explore how data scientists are able to share their work confidently with the right people. We will look at important concepts related to data and results sharing, quality assurance and data access restrictions.

Bringing DevOps practices and agile mindset to clinical reporting

In this module we explore how to make the most out of data science by developing the best mindset.

Version control and git flows for reproducible clinical reporting

In this module we introduce the idea of version control, and git in particular. We show how you can use git effectively to manage your code during clinical reporting, and how it can be used as a tool for collaboration. We also look at making an R project in particular reproducible

Making code reusable and robust in clinical reporting — a call for InnerSourcing

In this module we will discuss benefits of InnerSourcing, OpenSourcing and developing our own R packages. We will review some of the core principles and tools of R package development. Finally, we will learn how to set up a CI/CD workflow for R package development.

Assessing and managing risk

In this module we will review the tools and approaches used to understand risk in a codebase used to derive datasets and insights. By the completion of this module you will get some hands on experience applying these principles against a specific open source library.