Machine Learning

Machine learning is a type of artificial intelligence where a system learns from a set of data and finds patterns that can be applied to new data. Typically, machine learning requires having a set of data for training and a set of data for validation before it can be used accurately on new data.

In ME/CFS and related conditions, machine learning is perhaps most commonly used for finding patterns in big sets of data, like proteomics and metabolomics, that have the potential to become a biological signature of the disease. In these cases, the ultimate goal is to develop a diagnostic tool.

The Melbourne ME/CFS Collaboration put this technique to use in trying to distinguish ME/CFS from comorbid conditions. In this project, they developed a machine learning algorithm, trained on UK Biobank data, that can identify ME/CFS using a panel of metabolic markers and survey data with 83% accuracy. Read more about this study here.