Data Science in Clinical Practice

Careful data curation and analyses are essential in developing machine learning algorithms, that may usefully contribute to solving problems encountered in routine healthcare. Nevertheless, many valuable contributions never transition from the computer to the bedside. Often implementation is never attempted, or they fail to get the relevant CE marking (or equivalent local standard), or their implementation fails to elicit the intended health benefit (failure due to lack of clinical utility).

The current courses focus on the latter and provides the foundation necessary to plan and conduct clinical evaluation of machine learning solutions to fairly assess their contribution to clinical practice.