Subject clustering method and apparatus

A processor arrangement adapted to cluster similar subjects of a plurality of subjects, the processor arrangement comprising:

• an obtaining unit adapted to obtain a dataset of subject data, the dataset containing a respective set of quantitative values for each subject, each quantitative value being a measurement of a clinical or pathological feature of the subject and each set of quantitative values having a same number of quantitative values;
• and
• a clustering unit adapted to perform a clustering process by:
• performing a principal component analysis on the dataset of subject data to identify at least two principal components, each principal component being associated with a component score and a dataset of loadings, the number of loadings in each dataset of loadings being equal to the number of quantitative values in a set of quantitative values;
• generating a first dataset of weighting values, the number of weighting values in the first dataset being equal to the number of loadings in a dataset of loadings, wherein each weighting value in the first dataset of weighting values is based on a respective loading in the dataset of loadings associated with one of the first two principal components;
• weighting the quantitative values using the first dataset of weighting values to obtain a weighted dataset of subject data;
• and
• performing a clustering algorithm on the weighted dataset of subject data so as to cluster similar subjects into groups of subjects having similar clinical characteristics, wherein subjects in each group of subjects have a similar diagnosis, predicted subject outcome and/or suitable treatment options.