Implementation of an Adaptive Artificial Neural Network with Fuzzy Expert System for Diagnoses the Breast and Prostate Cancer: A Hybrid Technique

Abstract

Expert systems for medical applications have emerged as a result of recent developments in artificial intelligence. Additionally, in the recent years, computational tools have been developed to enhance the knowledge and skills of doctors when it comes to making decisions regarding their patients. The second-leading cause of cancer-related death is breast cancer, which is the most prevalent malignancy in women. About one-third of women with breast cancer pass away from the condition, even though it is treatable when found early. One of the main causes of death in the globe is cancer. Prostate cancer is one type of cancer that claims lives among men. This study is to evaluate the model's accuracy to expert forecasts regarding prostate cancer. Based on patient prostate volume, age, and prostate specific antigen data, predictions are made. Due to the absence of a prostate-like appearance in women, this disease mainly affects men. However, it might be challenging to distinguish between benign and malignant mammographic results, therefore in this work, we designed an expert system for Diagnosis of Breast Cancer and Prostate Cancer.  The findings demonstrate that the proposed fuzzy model can be used effectively to aid in the diagnosis and analysis of the possibility of prostate cancer and is one of the factors that doctors consider when determining whether or not a biopsy is necessary for these patients. The PCR value provided by the fuzzy model is within the PCR interval predicted by a specialist doctor. This technique makes it possible to avoid needless biopsy. Additionally, this technique can be useful for training medical students.