PURPOSE: An artificial neural network was developed to improve the prediction of pathological stage before radical prostatectomy based on variables available at biopsy and clinical parameters. MATERIALS AND METHODS: We used the prospectively accrued European prostate cancer detection data base to train an artificial neural network to predict pathological stage in 200 men with serum prostate specific antigen (PSA) 10 ng./ml. or less who underwent radical prostatectomy. Variables included in the artificial neural network were patient age, serum PSA, free-to-total PSA ratio, PSA velocity, transrectal ultrasound calculated total and transition zone volumes with their associated PSA parameters (transition zone PSA density and PSA density), digital rectal examination and Gleason score on biopsy. Two multilayer perceptron neural networks were trained on the remaining variables. Data on the 200 patients were divided randomly into a training set, a test set and a validation or prospective set. RESULTS: Overall classification accuracy of the artificial neural network was 92.7% and 84.2% for organ confined and advanced prostate cancer staging, respectively. For preoperatively predicting local versus advanced stage the area under the ROC curve for the artificial neural network was significantly larger (0.91) compared with logistic regression analysis (0.83), Gleason score (0.69), PSA density (0.68), prostate transition zone volume (0.63) and serum PSA (0.62) (all p <0.01). CONCLUSIONS: The artificial neural network outperformed logistic regression analysis and correctly predicted pathological stage in more than 90% of the validation patients with serum PSA 10 ng./ml. or less based on clinical, biochemical and biopsy data. In the future artificial neural networks may represent a significant step for improved staging of prostate cancer when counseling patients referred for radical prostatectomy or other curative treatments.
PURPOSE: An artificial neural network was developed to improve the prediction of pathological stage before radical prostatectomy based on variables available at biopsy and clinical parameters. MATERIALS AND METHODS: We used the prospectively accrued European prostate cancer detection data base to train an artificial neural network to predict pathological stage in 200 men with serum prostate specific antigen (PSA) 10 ng./ml. or less who underwent radical prostatectomy. Variables included in the artificial neural network were patient age, serum PSA, free-to-total PSA ratio, PSA velocity, transrectal ultrasound calculated total and transition zone volumes with their associated PSA parameters (transition zone PSA density and PSA density), digital rectal examination and Gleason score on biopsy. Two multilayer perceptron neural networks were trained on the remaining variables. Data on the 200 patients were divided randomly into a training set, a test set and a validation or prospective set. RESULTS: Overall classification accuracy of the artificial neural network was 92.7% and 84.2% for organ confined and advanced prostate cancer staging, respectively. For preoperatively predicting local versus advanced stage the area under the ROC curve for the artificial neural network was significantly larger (0.91) compared with logistic regression analysis (0.83), Gleason score (0.69), PSA density (0.68), prostate transition zone volume (0.63) and serum PSA (0.62) (all p <0.01). CONCLUSIONS: The artificial neural network outperformed logistic regression analysis and correctly predicted pathological stage in more than 90% of the validation patients with serum PSA 10 ng./ml. or less based on clinical, biochemical and biopsy data. In the future artificial neural networks may represent a significant step for improved staging of prostate cancer when counseling patients referred for radical prostatectomy or other curative treatments.
Authors: R P Meijer; E F A Gemen; I E W van Onna; J C van der Linden; H P Beerlage; G C M Kusters Journal: World J Urol Date: 2009-06-28 Impact factor: 4.226