Inteligência computacional aplicada à detecção de câncer de mama

Abstract

Breast cancer has a high death rate worldwide, and the most frequent in women, its diagnosis having beenperformed through screening, breast ultrasound and mammograms. This work aims to develop a classi er toidentify breast cancer using only anthropometric data and some parameters of a simple routine blood test thatare the biomarkers.The MLP Neural Networks and Neuro-Fuzzy Networks (ANFIS) were used for a decisioncommittee. This work demonstrates a breakthrough in the area of computational intelligence due to the goodresult of its classi cation of breast cancer, which was 97% accurate, a higher value presented compared tothe works of the last years that used similar biomarkers in the period of 2013 to the start of the year 2018.