Finding Temporal Patterns in Noisy Longitudinal Data: A Study in Diabetic Retinopathy

This paper describes an approach to temporal pattern mining using the concept of user dened temporal prototypes to dene the nature of the trends of interests. The temporal patterns are dened in terms of sequences of support values associated with identied frequent patterns. The prototypes are dened mathematically so that they can be mapped onto the temporal patterns. The focus for the advocated temporal pattern mining process is a large longitudinal patient database collected as part of a diabetic retinopathy screening programme, The data set is, in itself, also of interest as it is very noisy (in common with other similar medical datasets) and does not feature a clear association between specic time stamps and subsets of the data. The diabetic retinopathy application, the data warehousing and cleaning process, and the frequent pattern mining procedure (together with the application of the prototype concept) are all described in the paper. An evaluation of the frequ...

• Type: Conference
• Year: 2010
• Where: INCDM
• Authors: Vassiliki Somaraki, Deborah Broadbent, Frans Coenen, Simon Harding
• Keywords: INCDM 2010, Temporal Pattern Mining, Trend Mining.