Books by Patricia Melin

This book presents different experimental results as evidence of the good results obtained compared with respect to conventional approaches and literature references based on fuzzy logic. The main objective of the present work is the generation of fuzzy diagnosis systems that offer competitive classifiers to be applied in diagnosis systems.

This book is focused on the use of intelligent techniques, such as fuzzy logic, neural networks and bio-inspired algorithms, and their application in medical diagnosis. The main idea is that the proposed method may be able to adapt to medical diagnosis problems in different possible areas of the medicine and help to have an improvement in diagnosis accuracy considering a clinical monitoring of 24 hours or more of the patient.In this book, tests were made with different architectures proposed in the different modules of the proposed model. First, it was possible to obtain the architecture of the fuzzy classifiers for the level of blood pressure and for the pressure load, and these were optimized with the different bio-inspired algorithms (Genetic Algorithm and Chicken Swarm Optimization). Secondly, we tested with a local database of 300 patients and good results were obtained. It is worth mentioning that this book is an important part of the proposed generalmodel; for this reason, we consider that these modules have a good performance in a particular way, but it is advisable to perform more tests once the general model is completed.

In this book a new model for data classification was developed. Some architectures were developed in order to work mainly with two datasets, an arrhythmia dataset (using ECG signals) for classifying 15 different types of arrhythmias, and a satellite images segments dataset used for classifying six different types of soil.

This book focuses on the fields of hybrid intelligent systems based on fuzzy systems, neural networks, bio-inspired algorithms and time series. Prediction errors are evaluated by the following metrics: Mean Absolute Error, Mean Square Error, Root Mean Square Error, Mean Percentage Error and Mean Absolute Percentage Error.

The new approach was implement-ed as a hybrid intelligent system combining modular neural net-works and fuzzy systems. Finally, two genetic algo-rithms are used to perform the optimization of the modular neural networks parameters and fuzzy inference system parameters.