ABSTRACT
Artificial neural networks (ANNs) are algorithms that can be trained to recognise complex patterns in data sets. This chapter assesses the ability of an ANN to predict bladder cancer recurrence and stage progression in a group of patients with newly diagnosed Ta/T1 bladder cancer, and 12-month cancer-specific survival in a group of patients with primary T2-T4 bladder cancer using clinicopathological and molecular prognostic indicators. It demonstrates that using immunohisto-chemistry of frozen tumour sections, strong staining for epidermal growth factor receptor (EGFR) is associated with recurrence, progression, and reduced survival. EGFR status was confirmed to be an independent predictor of stage progression in Ta/T1 bladder cancer and survival in the group as a whole. In managing patients with bladder cancer, the principal problems for the clinician centre on the prompt diagnosis of bladder cancer, predicting which patients are at highest risk of tumour recurrence and stage progression in the Ta/T1 group and the relatively poor cancer-specific survival in the muscle-invasive group.
