Optimisation of a Machine Learning algorithm applied to detection of cancer-associated fibroblasts

Abstract

The activation of certain cells, such as fibroblasts, with the appearance of cancer has been studied and established. In order to accelerate and cheapen breast cancer diagnosis the optimisation of machine learning algorithms can be a powerful tool. Using two different data sets we studied the performance of different machine learning algorithms. We also studied the relevance of the different studied parameters, which could lead to some relevant biomedical conclusions. Further, we discuss the results obtained and discuss key aspects for improving the analysis of such experiments, as well as future directions for the use of Machine Learning in research against cancer