Preface

Machine learning in healthcare is a growing area of application of artificial intelligence in medicine. It is used in many areas covering classification and prediction problems. Artificial intelligence can be considered as systems or machines that aim to imitate the cognitive functions of people and improve themselves iteratively with the information they collect. The classification methods in the machine learning field, which is quite popular among artificial intelligence methods, can be used especially for various health data. In general, there are two types of classification approaches. The first is the binary classification approach, which sets the class tags as 0 or 1. The second is the methods that not only identify class labels but also determine class possibilities. The most prominent method for the first approach is the support vector machines method. Artificial neural networks, k-nearest neighbors, decision trees, and logistic regression methods are the methods included in the second approach. The logistic regression method is one of the most prominent methods among these methods. The logistic regression method performs the classification task by determining which class the data belongs to. The fact that this probability is close to 1 indicates that it increases the probability of being included in the related class and that it is close to 0 decreases the probability. The logistic regression method is used for early detection, diagnosis, and treatment in the field of health, from radiology to cancer, from neurology to cardiology, as well as outcome prediction and prognosis evaluation.

Mathematical modeling is used to have a better understanding and sometimes even to predict a pattern and results of the biological studies, and like all others, biological studies are no exception to this fact. Different methods of modeling are used dependent on the study and raw data available. The book will also contain several practical applications of how decision-making theory could be used in solving problems related to the selection of the best alternatives. In addition to machine learning, the book will focus on assisting decision-makers (government, organizations, companies, the general public, etc.) in making the best and most appropriate decision when confronted with multiple alternatives. The purpose of the analytical MCDM techniques is to support decision-makers under uncertainty and conflicting criteria while making a logical decision. Finally, the detail provided in the book will be of great help to the general public in their day-to-day life. The knowledge of the alternatives of the real-life problems, properties of their parameters, and the priority given to the parameters have a great effect on the consequences of the decisions. In this book, the application of MCDM has been provided for the real-life problems that occurred in health and biomedical engineering issues. In addition, the application of MCDM examples will be shown manually to users.