Introduction to Machine Learning in Medicine

Introduction of Machine Learning is certainly a revolutionary outbreak, both in the medical field and in many other research fields. The opportunities ML opens up are many, and it can be used to analyze medical images, to relate clinical data, to identify lesions, to support diagnosis, and to make predictions on prognosis and disease evolution, but it must be used carefully and competently so as not to run into wrong applications and consequently in misleading results. Machine Learning is represented by a series of algorithms which, through a learning process, are able to develop strategies to analyse, classify, and evaluate many types of data. However, there are important points on which to pay close attention, such as the preparation of the learning dataset, the choice of the most suitable algorithm for the aimed task, the evaluation of the results, and the weighing of the computing resources available. There are also some aspects related to Artificial Intelligence that need to be carefully considered, such as evaluation of the developed model, explainability, interpretability, reproducibility of the outputs, and all the legal and ethical issues that the nature of these approaches generates. Taking the necessary precautions for use, however, it cannot be denied that the application of ML in the medical field opens up very interesting and revolutionary perspectives, especially in the field of precision and personalized medicine.