Knowledge representation for lung cancer patients' prognosis

The gradual increase of cancer cases worldwide has been posing a need on the use of computing resources to accurately retrieve the information recorded in databases. One can highlight the retrieved information importance from a specialist in order to better evaluate pathological response and predict the cancer patient prognosis. This paper presents a way to represent knowledge of cancer registries with emphasis on prognosis. It makes use data mining techniques to find patterns in data stored for patient's lifetime in similar situations. The work is focused on the generation of association rules to find patterns on these registries in order to measure the patient prognosis and drive healthcare experts conclusions. A validation against international oncology organizations and health publications was also made to ensure data and work reliability.