Giannopoulou E.G. (ed.) Data Mining in Medical and Biological Research

InTech, 2008, -332 p.Data mining is the research area involving powerful processes and tools that allow an effective analysis and exploration of usually large amounts of data. In particular, data mining techniques have found application in numerous different scientific fields with the aim of discovering previously unknown patterns and correlations, as well as predicting trends and behaviors.
In the meantime, during the current century of biomedical sciences, biology and medicine have undergone tremendous advances in their technologies and therefore have generated huge amounts of biomedical information. This book intends to bring together the most recent advances and applications of data mining research in the promising areas of medicine and biology from around the world. Seventeen chapters, twelve related to medical research and five focused on the biological domain, describe interesting applications, motivating progress and worthwhile results.
Chapter 1 presents a novel concept of improving the classification accuracy for common learning algorithms by uncovering the potential relevance of features through multivariate interdependent attribute preprocessing methods.
Chapter 2 introduces a hybrid approach, combining discrete wavelet transform and neural networks, for the classification of a complex dataset such as an electroencephalography time series and claims that a high degree of classification accuracy can be achieved.
Chapter 3 explores the possibility that DNA viruses play an important role in development of breast tumors, using artificial neural networks and agglomerative hierarchical clustering techniques.
Chapter 4 investigates the research hypothesis of developing certain domain ontology from discovered rules. To practically examine this assumption, the proposed approach integrates different data mining techniques to assist in developing a set of representative consensual concepts of the underlying domain.
Chapter 5 describes the process of medical knowledge acquisition and outlines the requirements for producing knowledge that can be trusted and applied in a clinical setting. This work demonstrates that the individual needs of medical professionals can be addressed and that data mining can be a valuable tool in the diagnostic and decision making toolkit.
Chapter 6 provides an overview of the various technologies in a radiology department that allow data mining, and describes case-examples utilizing data mining technologies.
Chapter 7 demonstrates the results of a nationwide study that applied data and text mining techniques to medical near-miss case data related to medicines and medical equipments. This study also analyzed the causal relationship of occurrences of the near-miss
cases and the opinions on the counter measures against them.
Chapter 8 presents a complex clinical problem that has been addressed using particle swarm optimization, a suitable approach for data mining subject to a rule base which defines the quality of rules and constancy with previous observations.
Chapter 9 focuses on the description of a model for monitoring and planning of human resources in a nationwide public health-care system. The aim of the presented monitoring system is to assess performance and provide information for the management of a primary health-care network.
Chapter 10 summarizes the core technologies and the current research activities regarding the interoperability of information and knowledge extracted from data mining operations.
Chapter 11 describes a series of sensory vital-sign measurement technologies to facilitate data collection in the daily environment, and the application of data mining algorithms for the interpretation of comprehensive large-scale data.
Chapter 12 describes and suggests a mobile phone based intelligent electrocardiogram signal telemonitoring system that incorporates data mining and knowledge management techniques.
Chapter 13 suggests that using statistical measures of independence can indicate nonsignificant DNA motifs, whereas measures of significance based on shape distributions can be extremely informative.
Chapter 14 offers an explanatory review on the post genomic field and demonstrates representative and outstanding examples that highlight the importance of data mining methods in proteomics research.
Chapter 15 presents a variety of methods for a standard graph analysis, particularly oriented to the study of cellular networks, and indicates that the network approach provides a suitable framework for exploring the organization of biomolecules.
Chapter 16 demonstrates a framework for biomedical information extraction from text, which integrates a data mining module for extraction rule discovery.
Chapter 17 describes the attempt to exploit social data mining in order to improve results in bio-inspired intelligent systems, using a plethora of data mining techniques.