Article Text
Abstract
Introduction/Background The diagnostic accuracy of artificial neural networks (ANNs) and Classification and Regression Trees (CARTs) has been already tested in several studies involving cancer patients. To date, however, its use in the field of gynecologic oncology remains extremely limited. In our study, we aim to investigate the diagnostic accuracy of three different methodologies logistic regression, ANNs and CARTs for the prediction of endometrial cancer in post-menopausal women.
Methodology We retrospectively searched medical records of all patients that were subjected in endometrial sampling procedures due to post-menopausal vaginal bleeding or ultrasographic evidence of endometrial pathology. Classical regression analysis was performed along with ANN and CART analysis using the IBM SPSS and Matlab statistical packages.
Results Overall, 178 women were enrolled. Among them, 106 women were diagnosed with carcinoma; whereas the remaining 72 women had normal histology in the final specimen. ANN analysis seems to perform better with a Sensitivity of 86.8%, Specificity of 83.3% and overall accuracy of 85.4%. CART analysis did not perform as well with a Sensitivity of 78.3% a Specificity of 76.4% and overall accuracy of 77.5%. Regression analysis had a poorer predictive accuracy with a sensitivity of 76.4%, a specificity of 66.7% and an overall accuracy of 72.5%.
Conclusion Artificial intelligence is a powerful mathematical tool that may guide clinicians involved in primary care in decision making when endometrial pathology is suspected. Our study clearly denotes that this can be done with simple indices which may be drawn from the patients` history and a basic anthropometric measurement. However, current data in this field remain in a premature state to support this method as a screening tool as the number of participants remains low to stabilize the diagnostic accuracy of the algorithm; hence, future, multicenter studies are needed to support our findings.
Disclosure Nothing to disclose.