Neural network analysis of clinicopathological factors in urological disease: a critical evaluation of available techniques

A comparison of the performance of microsatellite and methylation urine analysis for predicting the recurrence of urothelial cell carcinoma, and definition of a set of markers by Bayesian network analysis

OBJECTIVE

To compare the potential of two diagnostic methods for detecting recurrence of urothelial cell carcinoma (UCC) of the bladder, by (i) detecting alterations in microsatellite DNA markers and loss of heterozygosity (LOH), and (ii) detecting aberrant gene hypermethylation, as UCC has a high recurrence rate in the urinary tract and the disease can invade muscle if new tumours are overlooked.

PATIENTS AND METHODS

Over 1 year, urine samples were retrieved from 40 patients already diagnosed with bladder UCC (30 pTa, two pTis, eight pT1). Samples were collected 6 months after bladder tumour resection, during the follow-up schedule. We used samples to analyse nine microsatellite markers and the methylation status of 11 gene promoters. Receiver operating characteristic curves were generated and Bayesian statistics used to create an interaction network between recurrence and the biomarkers.

RESULTS

During the study, 15 of the 40 patients (38%) had a tumour recurrence and 14 were identified by cystoscopy (reference method). Overall, microsatellite markers (area under curve, AUC 0.819, 95% confidence interval, CI, 0.677-0.961) had better performance characteristics than promoter hypermethylation (AUC 0.448, 0.259-0.637) for detecting recurrence. A marker panel of IFNA, MBP, ACTBP2, D9S162 and of RASSF1A, and WIF1 generated a higher diagnostic accuracy of 86% (AUC 0.92, 0.772-0.981).

CONCLUSION

Microsatellite markers have better performance characteristics than promoter hypermethylation for detecting UCC recurrence. These data support the further development of a combination of only six markers from both methods in urinary DNA. Once validated, it could be used routinely during the follow-up for the early detection and surveillance of UCC from the lower and upper urinary tract.

Prediction of survival in surgical unresectable lung cancer by artificial neural networks including genetic polymorphisms and clinical parameters

Lung cancer, a common malignancy in Taiwan, involves multiple factors, including genetics and environmental factors. The survival time is very short once cancer is diagnosed as being in advanced stage and surgically unresectable. Therefore, a good model of prediction of disease outcome is important for a treatment plan. We investigated the survival time in advanced lung cancer by using computer science from the genetic polymorphism of the p21 and p53 genes in conjunction with patients' general data. We studied 75 advanced and surgical unresectable lung cancer patients. The prediction of survival time was made by comparing real data obtained from follow-up periods with data generated by an artificial neural network (ANN). The most important input variable was the clinical staging of lung cancer patients. The second and third most important variables were pathological type and responsiveness to treatment, respectively. There were 25 neurons in the input layer, four neurons in the hidden layer-1, and one neuron in the output layer. The predicted accuracy was 86.2%. The average survival time was 12.44 +/- 7.95 months according to real data and 13.16 +/- 1.77 months based on the ANN results. ANN provides good prediction results when clinical parameters and genetic polymorphisms are considered in the model. It is possible to use computer science to integrate the genetic polymorphisms and clinical parameters in the prediction of disease outcome. Data mining provides a promising approach to the study of genetic markers for advanced lung cancer.

The use of neural networks for predicting the result of endoscopic treatment for vesico-ureteric reflux

OBJECTIVE

To create an artificial neural network (ANN) to aid in predicting the results of endoscopic treatment for vesico-ureteric reflux (VUR).

MATERIALS AND METHODS

During 1999-2001 we used endoscopic treatment in 261 ureteric units with VUR of all grades and causes. An ANN based on multilayer perceptron architecture was created using an 11 x 6 x 1 structure, taking the following as variables: the cause and grade of VUR, the patient's age and sex, the type of implanted substance and its volume, the number of treatments, the affected ureter, the endoscopic findings, and the type of cystography used. In all, 174 cases were used as training samples for the ANN and 87 to validate it. We calculated the sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and the success rate (%) of the system.

RESULTS

In the training group the ANN gave a sensitivity of 86.4%, a specificity of 89.5%, a PPV of 76% and NPV of 94%, with a success rate of 88.6%. In the same training group logistic regression (LR) gave respective values of 68.2%, 58.8%, 39%, 82.7% and 61.4%. In the validation group the respective values for the ANN were 71.4%, 81.6%, 58.8%, 88.6% and 78.9%, and in the same validation group the LR gave 64.4%, 50%, 32.1%, 79.2% and 53.9%. The Wilcoxon test confirmed the independence of both methods (P < 0.001).

CONCLUSION

The ANN is an effective tool for assisting the urologist in indicating and applying endoscopic treatments for VUR.

Mass spectrometry as a diagnostic and a cancer biomarker discovery tool: opportunities and potential limitations

Serum proteomic profiling, by using surfaced-enhanced laser desorption/ionization-time-of-flight mass spectrometry, is one of the most promising new approaches for cancer diagnostics. Exceptional sensitivities and specificities have been reported for some cancer types such as prostate, ovarian, breast, and bladder cancers. These sensitivities/specificities are far superior to those obtained by using classical cancer biomarkers. In this review, I concentrate more on questions that cast doubt on the results reported and propose experiments to investigate these questions in detail, before the technique is used at the clinic. It is clear that the method needs to be externally and thoroughly validated before clinical implementation is warranted.

Prediction of stone disease by discriminant analysis and artificial neural networks in genetic polymorphisms: a new method

OBJECTIVE

To use information from genetic polymorphisms and from patients (drinking/exercise habits) to identify their association with stone disease, the main analytical and predictive tools being discriminant analysis (DA) and artificial neural networks (ANNs).

PATIENTS, SUBJECTS AND METHODS

Urinary stone disease is common in Taiwan; the formation of calcium oxalate stone is reportedly associated with genetic polymorphisms but there are many of these. Genotyping requires many individuals and markers because of the complexity of gene-gene and gene-environmental factor interactions. With the development of artificial intelligence, data-mining tools like ANNs can be used to derive more from patient data in predicting disease. Thus we compared 151 patients with calcium oxalate stones and 105 healthy controls for the presence of four genetic polymorphisms; cytochrome p450c17, E-cadherin, urokinase and vascular endothelial growth factor (VEGF). Information about environmental factors, e.g. water, milk and coffee consumption, and outdoor activities, was also collected. Stepwise DA and ANNs were used as classification methods to obtain an effective discriminant model.

RESULTS

With only the genetic variables, DA successfully classified 64% of the participants, but when all related factors (gene and environmental factors) were considered simultaneously, stepwise DA was successful in classifying 74%. The results for DA were best when six variables (sex, VEGF, stone number, coffee, milk, outdoor activities), found by iterative selection, were used. The ANN successfully classified 89% of participants and was better than DA when considering all factors in the model. A sensitivity analysis of the input parameters for ANN was conducted after the ANN program was trained; the most important inputs affecting stone disease were genetic (VEGF), while the second and third were water and milk consumption.

CONCLUSIONS

While data-mining tools such as DA and ANN both provide accurate results for assessing genetic markers of calcium stone disease, the ANN provides a better prediction than the DA, especially when considering all (genetic and environmental) related factors simultaneously. This model provides a new way to study stone disease in combination with genetic polymorphisms and environmental factors.

Estimation of prostate cancer risk on the basis of total and free prostate-specific antigen, prostate volume and digital rectal examination

BACKGROUND AND OBJECTIVE

Approximately 70% of the men with an elevated serum prostate-specific antigen (PSA) identified in prostate cancer screening do not have prostate cancer. Other available diagnostic variables may be utilized to reduce the number of false positive PSA results, but few algorithms for calculation of the combined impact of multiple variables are available. The objective of this study was to establish nomograms showing the probability of detecting prostate cancer at biopsy on the basis of total PSA, and the percentage of free PSA in serum, prostate volume and digital rectal examination (DRE) findings.

METHODS

In a randomized, population-based prostate cancer screening trial 10284 men aged 55-67 years were screened during 1996 and 1997 in two metropolitan areas in Finland. Results for men (n=758) with a serum PSA of 4-20 microg/l were used to establish the risk nomograms. Of these 200 (26%) had prostate cancer at biopsy.

RESULTS

Prostate cancer probability depended most strongly on the percentage of free PSA. Total PSA, prostate volume, and DRE also contributed to prostate cancer probability, whereas age and family history of prostate cancer did not. More false positive PSA results could be eliminated by using the multivariate risk model rather than the percentage of free PSA (p<0.001) or PSA density (p=0.003) alone.

CONCLUSIONS

Wide variation in probability of detecting prostate cancer among screened men with a serum PSA of 4-20 microg/l was observed. The nomograms established can be used to avoid or defer biopsy in men with a low prostate cancer probability in spite of a serum PSA level exceeding 4 microg/l.