Find in Library

Introduction: Breast cancer is a malignant tumor in the breast tissue cells caused by genetic disorders where the tumor cells start dividing and proliferating without any control. Cancer is associated with increasing the surface of pro-inflammatory cytokines, such as interleukin 6 (IL6), interleukin 8 (IL8), and vascular endothelial growth factor (VEGF). Coenzyme Q10 is a natural compound that performs anti-inflammatory functions by decreasing the secretion of cytokines. Objective: To develop a diagnostic system for persons with breast cancer, and to predict the impact levels of Q10 supplementation on pro-inflammatory factors using an artificial neural network (ANN) and the logistic regression method. Method: The dataset was created based on a study on the subjects referred to the Medical Oncology Department of the Governmental Hospital of Ahvaz, Iran. The extracted properties including the surface of IL6, IL8, and VEGF were used as the main research factors in the modeling development. A model for diagnosing the patients with breast cancer was designed. Moreover, a predictor system was developed for estimating the impacts of Q10 supplementation on pro-inflammatory factors. Moreover, the results obtained from the logistic regression method based on the input pro-inflammatory variables were compared with those of the proposed ANN predictor model. Results: The results showed that the accuracy of the ANN model in diagnosing breast cancer was 79.8%, while the accuracy for predicating the Q10 impact levels on each inflammatory factor, i.e., IL6, IL8, and VEGF, was 96%, 88%, and 92%, respectively. On the other hand, the accuracy of the logistic regression technique for predicting the impact of Q10 on IL6, IL8, and VEGF levels was 70%, 66%, 80%, respectively. Conclusion: In this study, the utilized ANN technique had a relatively high sensitivity and precision for identifying people with breast cancer. In addition, predicting the effects of Q10 on inflammatory factors using the ANN provided a high accuracy compared to the logistic regression method. The results of research on artificial intelligence can have a great impact on physicians’ decisions to predict the effects of Q10 supplementation on improving the patients in their specific conditions.