Colorectal cancer has risen in incidence to become the second commones
t form of cancer in Singapore. The primary treatment is surgery but up
to 50% of patients still suffer from recurrence of the cancer after s
urgery. Early identification of recurrence will increase the effective
ness of therapy and the survival of patients. This paper describes the
CARES (Cancer Recurrence Support) System, whose objective is to predi
ct the recurrence of colorectal cancer, using Case-based Reasoning (CB
R), and supported by other techniques such as data mining and natural
language processing. The CARES System employs CBR to compare and contr
ast between the new and past colorectal cancer patient cases, and make
s inferences based on those comparisons to determine the high risk pat
ient groups. The features and functionality of the system are describe
d. (C) 1997 Elsevier Science B.V.