Prof. Patricia Melin: “Hybrid Intelligent Systems in Medical Diagnosis”

Abstract

Hybrid intelligent systems are formed by prudent combinations of intelligent models, such as neural networks, fuzzy models and others, to achieve efficient solutions to real-world problems. The main idea is to take advantage of the main characteristics of the individual models. For example, neural networks are good for learning from training data, while fuzzy logic is good for representing expert knowledge and uncertainty management, and evolutionary computing is good for search and optimization. Medical diagnosis are challenging due to their complexity and the uncertainty involved in the inherent decision making process done by Medical Doctors. In our work the proposed approach is to build powerful hybrid intelligent systems for achieving the automated medical diagnosis. The proposed hybrid architecture is based on modular neural networks for learning form large datasets of patients. Then for combining the outputs of the modules an integration based on type-2 fuzzy rules is performed for modeling the involved decision making process, as well as the inherent uncertainty in making the decisions. Finally, evolutionary or bio-inspired optimization techniques are used for optimizing the architectures of the neural networks, as well as the structures of the type-2 fuzzy systems. The proposed hybrid architecture has been tested and applied in different medical diagnosis problems with good results. In particular, we can mention hypertension diagnosis, arrhythmia diagnosis and diabetes diagnosis and detection of pulmonary diseases. We believe that the proposed hybrid intelligent approach can also be used for other diagnosis problem in the future.

Curriculum Vitae

Prof. Patricia Melin holds the Doctor in Science degree (Doctor Habilitatus, D.Sc.) in Computer Science from the Polish Academy of Sciences. She is a Professor of Computer Science in the Graduate Division, Tijuana Institute of Technology, Tijuana, Mexico, since 1998. In addition, she is serving as Director of Graduate Studies in Computer Science and is Head of the research group on Hybrid Neural Intelligent Systems (2000-present). Prof. Melin has published nearly 800 publications in indexed journals, book chapters, and conference proceedings, as well as nearly 50 books, and as consequence of this she has achieved more than 15000 citations with an H-index of 66 in Google Scholar. In addition, she has been awarded the Highly Cited Researcher recognition in the area of Computer Science in 2017 and 2018 by Clarivate Analytics, being in the top 1% cited authors in this area. She has also been advisor of more than 85 graduate students in computer science at the Ph.D. and masters levels.
She has been the President of NAFIPS (North American Fuzzy Information Processing Society) in 2019-2020. Prof. Melin is the founding Chair of the Mexican Chapter of the IEEE Computational Intelligence Society. She is member of the IEEE Neural Network Technical Committee (2007-present), the IEEE Fuzzy System Technical Committee (2014 to present) and is Chair of the Task Force on Hybrid Intelligent Systems (2007-present) and she is currently Associate Editor of the Information Sciences Journal, IEEE Transactions on Fuzzy Systems and Journal of Complex and Intelligent Systems. She is a member of NAFIPS, IFSA, and IEEE. She belongs to the Mexican Research System with level III (highest level). Her research interests are in Modular Neural Networks, Type-2 Fuzzy Logic, Pattern Recognition, Fuzzy Control, Neuro-Fuzzy and Genetic-Fuzzy hybrid approaches. She has served as Guest Editor of several Special Issues in the past, in journals like: Applied Soft Computing, Intelligent Systems, Information Sciences, Non-Linear Studies, Engineering Applications of Artificial Intelligence, Fuzzy Sets and Systems.