Breast Cancer Early Detection Comparison with Deep Learning and Machine Learning Models: A Case of Study

Breast cancer is one of the most widespread in the female population, being able to predict its developments and capturing the inputs of the onset of the disease is one of the main objectives that science is pursuing. Clinical Decision Support Systems (CDSS) in recent decades are extensively using these technological tools, such as Machine Learning (ML) and Deep Learning (DL). In this paper, two of the main methods of these subset of AI are compared: an ensemble-type algorithm, XGBoost (or Extreme Gradient Boosting) and a deep neural network (DNN) are applied to the data of a study conducted on an Indonesian population. The results obtained are very interesting as despite being tabular, binary categorical and multiclass data, the DNN model achieves performance and results much higher than the well-known XGB used in literature for data of this type.