Edinburgh Napier University

  Cancer treatments are now more effective than ever and, as a consequence, cancer is becoming a chronic disease. Chemotherapy is a frequently used treatment in people with cancer and it can cause a number of side-effects which if not properly managed could have a negative impact on the patients' quality of life. In this study, a sample of 56 patients receiving chemotherapy treatment for breast, colorectal and lung cancer is considered; each experienced side-effect is recorded during four consecutive treatment cycles (each lasting 14 days). Five of the most frequent side-effects (fatigue, nausea, mucositis, hand and foot sore, diarrhoea) are selected to build a comprehensive model which predicts the probability of experiencing a certain symptom on a specified day of each cycle of therapy. The computed accuracy of results shows that the newly proposed model has an enhanced predictive power compared to a state-of-the-art approach. The information gained from this study will help medical and nursing staff caring for such patients to more accurately predict the side-effects that patients will experience and therefore select appropriate help to minimise, whenever possible, the influence of those symptoms.