Proposta e avaliação de um modelo híbrido de seleção de características para o prognóstico do câncer de mama

Abstract

Artificial Intelligence technology has been instrumental in the role of health care for society. It has been widely used in various branches of medicine. One of its main applications is in the context of breast cancer disease prognosis. Cancer is considered the second leading cause of death from disease in the world. In this context, breast cancer stands out, which is considered the largest occurrence of cancer among women in the world. One of the main challenges in this scenario is to identify which are the most relevant characteristics in the development of this type of neoplasm by a patient. This type of filtering is performed by feature selection methods. This paper presents a hybrid feature selection model that should be used by clinicians of a patient in order to perform a breast cancer mortality prediction. The hybrid model is composed of two algorithms already existing in the literature: Gain Ratio and ReliefF. In comparison with the models already existing in the literature, the proposed algorithm presented better results for the evaluation metrics used for this purpose.