Neural network analysis of combined conventional and experimental prognostic markers in prostate cancer: a pilot study
1998

Using Neural Networks to Predict Prostate Cancer Outcomes

Sample size: 41 publication Evidence: moderate

Author Information

Author(s): RNG Naguib, MC Robinson, DE Neal, FC Hamdy

Primary Institution: University of Newcastle upon Tyne

Hypothesis

Can artificial neural networks improve the prediction of outcomes in prostate cancer compared to conventional statistical methods?

Conclusion

Artificial neural networks can accurately predict outcomes in prostate cancer by analyzing a combination of conventional and experimental prognostic markers.

Supporting Evidence

  • 80% of patients were correctly classified regarding outcome using the combination of factors.
  • Excluding both bcl-2 and p53 reduced prediction accuracy to 60%.
  • Statistical significance was attained when all markers were considered (P = 0.0096).

Takeaway

This study shows that computers can help doctors predict how prostate cancer will behave by looking at different factors together.

Methodology

The study used artificial neural networks to analyze data from 41 prostate cancer patients, comparing the predictions with conventional statistical methods.

Limitations

The study had a small sample size and was a pilot study, indicating that further research is needed.

Participant Demographics

Participants were 41 men aged 47 to 86 years with histologically proven prostate cancer.

Statistical Information

P-Value

0.0032

Confidence Interval

0.3585 < K < 0.9459

Statistical Significance

p<0.05

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication