Artificial intelligence has made great advances in the healthcare field, particularly in medical imaging. However, data and annotations in this area are often scarce and expensive. Although essential for machine learning models, labeling images is a tedious and time-consuming task. Active learning addresses this challenge by selecting informative samples, and using less annotated data while still getting a good model performance. The proposed solution uses the PatchCamelyon dataset, with patches from histopathologic scans of sentinel lymph node sections for the detection of metastatic tissue in breast cancer patients. This work proposes an active learning approach that uses both the informativeness of the sample and the distribution of data. The solution showed promising results when compared to a random sampling approach. In some iterations, the difference between the F1 score in the proposed active learning solution and random sampling was greater than 0,20.