Advances in new omics technologies, such as genomics, proteomics, transcriptomics, metabolomics, among others. They have made it possible to obtain huge amounts of data related to the functioning of biological processes. However, to understand the phenomenon that is happening in the development of a disease such as acute myeloid leukemia, it is not enough with a single type of data due to the high biological complexity, so it is necessary to integrate these to understand better what is happening. Approaches using machine learning models for data integration such as autoencoders are an alternative for the study of diseases. For this reason, this work aims to develop a survival classification model among acute myeloid leukemia patients using autoencoder described by Nikola Simidjievski et.al 2019 and compare the results with those obtained with the raw data and the data transformed by PCA.