Catálogo de publicaciones - libros

In this chapter we provide a broad overview of selected knowledge management, data mining, and text mining techniques and their use in various emerging biomedical applications. It aims to set the context for subsequent chapters. We first introduce five major paradigms for machine learning and data analysis including: probabilistic and statistical models, symbolic learning and rule induction, neural networks, evolution-based algorithms, and analytic learning and fuzzy logic. We also discuss their relevance and potential for biomedical research. Example applications of relevant knowledge management, data mining, and text mining research are then reviewed in order including: ontologies; knowledge management for health care, biomedical literature, heterogeneous databases, information visualization, and multimedia databases; and data and text mining for health care, literature, and biological data. We conclude the paper with discussions of privacy and confidentiality issues of relevance to biomedical data mining.

The ability to create a big picture of a knowledge domain is valuable to both experts and newcomers, who can use such a picture to orient themselves in the field’s intellectual space, track the dynamics of the field, or discover potential new areas of research. In this chapter we present an overview of medical informatics research by applying domain visualization techniques to literature and author citation data from the years 1994–2003. The data was gathered from NLM’s MEDLINE database and the ISI Science Citation Index, then analyzed using selected techniques including self-organizing maps and citation networks. The results of our survey reveal the emergence of dominant subtopics, prominent researchers, and the relationships among these researchers and subtopics over the ten-year period.

As biomedical research and healthcare continue to progress in the genomic/post genomic era a number of important challenges and opportunities exist in the broad area of biomedical informatics. In the context of this chapter we define bioinformatics as the field that focuses on information, data, and knowledge in the context of biological and biomedical research. The key challenges to bioinformatics essentially all relate to the current flood of raw data, aggregate information, and evolving knowledge arising from the study of the genome and its manifestation. In this chapter we first briefly review the source of this data. We then provide some informatics frameworks for organizing and thinking about challenges and opportunities in bioinformatics. We use then use one informatics framework to illustrate specific challenges from the informatics perspective. As a contrast we provide also an alternate perspective of the challenges and opportunities from the biological point of view. Both perspectives are then illustrated with case studies related to identifying and addressing challenges for bioinformatics in the real world.

This chapter explores issues in managing privacy and security of healthcare information used to mine data by reviewing their fundamentals, components and principles as well as relevant laws and regulations. It also presents a literature review on technical issues in privacy assurance and a case study illustrating some potential pitfalls in data mining of individually identifiable information. The chapter closes with recommendations for privacy and security good practices for medical data miners.

The development of modern bioethics has been strongly influenced by technology. Important ethical questions surround the use of electronic health records, clinical decision support systems, internet-based consumer health information, outcome measurement, and data mining. Electronic health records are changing the way health information is managed, but implementation is a difficult task in which social and cultural issues must be addressed. Advice produced by decision support systems must be understood and acted upon in the context of the overall goals and values of health care. Empowering health care consumers through readily-available health information is a valuable use of the internet, but the nature of the internet environment raises the spectre of abuse of vulnerable patients. Outcome studies have inherent value judgments that may be hidden. Data mining may impact confidentiality or lead to discrimination by identifying subgroups. All of these issues, and others, require careful examination as more and more health information is captured electronically.

The description of concepts in the biomedical domain spans levels of precision, complexity, implicit knowledge, and breadth of application that makes the knowledge representation problem more challenging than that in virtually any other domain. This chapter reviews some of this breadth in the form of use-cases, and highlights some of the challenges confronted, including variability among the properties of terminologies, classifications, and ontologies. Special challenges arise at the semantic boundary between information and terminology models, which are not resolvable on one side of either boundary. The problems of aggregation are considered, together with the requirement for rule-based logic when mapping information described using detailed terminologies to high-level classifications. Finally, the challenge of semantic interoperability, arguably the goal of all standards efforts, is explored with respect to medical concept representation.

The importance of biomedical concept relationships and document concept interrelationships are discussed and some of the ways in which concept relationships have been used in information search and retrieval are reviewed. We look at examples of innovative approaches utilizing biomedical concept identification and relationships for improved document and information retrieval and analysis that support knowledge creation and management.

Ontology design is an important aspect of medical informatics, and reusability is a key issue that is determined by the level of compatibility among ontology concepts and among the theories of the biomedical domain they convey. In this article, we examine OpenGALEN, the UMLS Semantic Network, SNOMED CT, the Foundational Model of Anatomy, and the MENELAS ontology as well as descriptions of the biomedical domain in two general ontologies, OpenCyc and WordNet. Using the representation of in each system, we examine issues in compatibility among these ontologies. The presence of additional knowledge is also illustrated and some issues in creating and aligning biomedical ontologies are discussed.

The field of information retrieval (IR) is generally concerned with the indexing and retrieval of knowledge-based information. Although the name implies the retrieval of any type of information, the field has traditionally focused on retrieval of text-based documents, reflecting the type of information that was initially available by this early application of computer use. However, with the growth of multimedia content, including images, video, and other types of information, IR has broadened considerably. The proliferation of IR systems and on-line content has also changed the notion of libraries, which have traditionally been viewed as buildings or organizations. However, the developments of the Internet and new models for publishing have challenged this notion as well, and new digital libraries have emerged.

Given the amount of literature relevant to many of the areas of biomedicine, researchers are forced to use methods other than simply reading all the literature on a topic. Necessarily one must fall back on some kind of search engine. While the Google PageRank algorithm works well for finding popular web sites, it seems clear one must take a different approach in searching for information needed at the cutting edge of research. Information which is key to solving a particular problem may never have been looked at by many people in the past, yet it may be crucial to present progress. What has worked well to meet this need is to rank documents by their probable relevance to a piece of text describing the information need (a query). Here we will describe a general model for how this is done and how this model has been realized in both the vector and language modeling approaches to document retrieval. This approach is quite broad and applicable to much more than biomedicine. We will also present three example document retrieval systems that are designed to take advantage of specific information resources in biomedicine in an attempt to improve on the general model. Current challenges and future prospects are also discussed.