ABSTRACT
Prostate cancer is the second most common malignancy in men in the United Kingdom with approximately 14,000 new cases diagnosed and 9,000 men dying from the disease every year. Deregulation of the genetic pathway leading to programmed cell death may confer hormone-resistant prostate cancer which is incurable. This chapter examines the ability of Artificial neural networks (ANNs) to assess novel prognostic markers, in addition to established clinical and biochemical parameters. The most commonly used criteria influencing clinical decision-making in treating prostate cancer are a combination of: patient’s age and life expectancy, tumour stage and grade, and serum prostate-specific antigen levels. In prostate cancer, studies have evaluated the use of ANNs in diagnosis and the prediction of recurrence following radical surgery. The results showed high sensitivity and specificity rates in predicting biopsy results in men with suspected prostate cancer, and recurrence following radical prostatectomy.
