It is interesting to watch the evolution and development of rapidly growing disciplines such as medical and health informatics. It took the seminal 1990 work by Shortliffe et al.  more than ten years to reach the need for a second edition, and then only six years to reach a third. Since the turn of the new millennium, there has been an explosion in the number and diversity of textbook-length works in medical informatics and related topics.
As a faculty member who enthusiastically embraced Shortliffe’s second edition when published, and used it in a doctoral course in health informatics for several years, I was initially thinking the recent work of Chen et al. might provide a replacement. Indeed, the editors feel the text has wide potential appeal: in senior undergraduate and graduate-level courses in medical informatics and health computing, computer science courses in biomedical computing and data mining, and library and information sciences and knowledge management courses. Chen’s work does not make Shortliffe’s third edition obsolete, however; it remains essential as a general introductory core text for medical informatics. Chen’s work does substantially cover a range of new topics, and, in many ways, is a complementary addition that picks up where the work by Shortliffe et al. left off. In turn, Shortliffe’s third edition has been renamed Biomedical informatics: computer applications in health care and biomedicine, reflecting the growth of informatics into such areas such as bioinformatics, health administration, nursing, and other allied health areas beyond the core medical disciplines.
Like the book by Shortliffe et al., Chen’s work is divided into three sections. The differences between the two works become immediately apparent, however. The core divergence is in the focus on data mining and knowledge management applications, which are not even covered in any of Shortliffe’s editions, mainly due to their immaturity and niche positioning in the field in 2000. Unit 1 in this book, “Foundational Topics in Medical Informatics,” is made up of five chapters, spanning 160 pages, including coverage of knowledge management, data mining and text mining, the mapping of medical informatics research, bioinformatics, information security and privacy, and the ethical and social challenges of electronic health information. By contrast, Shortliffe’s Unit 1, “Recurrent Themes in Medical Informatics,” provides a broader (11 chapters) and lengthier (446 pages) introduction to the discipline, introducing the discipline of medical informatics from a historical perspective, and discussing medical data and its uses, medical decision making, the essentials of medical computing, standards, natural language processing, ethical issues, and evaluation and assessment.
Unit 2 of Chen’s work, the eight-chapter, 236-page “Information and Knowledge Management,” covers medical concept representation, medical concept relationships, biomedical ontologies, information retrieval and digital libraries, text retrieval in biomedicine, anatomic and three-dimensional (3D) imaging, and infectious disease informatics. This compares with Shortliffe’s 11-chapter, 345-page Unit 2, “Biomedical Informatics Applications,” which covers computer-based patient record systems and management of information in the healthcare enterprise, consumer health informatics and telehealth, integrated delivery networks, patient care and monitoring systems, imaging and information retrieval, clinical decision support systems, and computing in medical education and bioinformatics.
Unit 3 of Chen’s work, “Text Mining and Data Mining,” spanning eight chapters and 225 pages, covers semantic interpretation for biomedical literature, semantic text parsing of patient records, identification of biological relationships from text documents, metabolic network modeling and visualization, gene pathway text mining and visualization, genomic data mining, exploratory genome analysis, and joint learning techniques using multiple data and knowledge types. In comparison, Shortliffe’s Unit 3, “Medical Informatics in the Years Ahead,” offers a two-chapter, 55-page view ahead to further developments in computer applications in healthcare.
Chen’s work provides some minor content overlap with Shortliffe in the areas of biomedical ontologies and concept representation, in the ethical considerations of the electronic patient record, and in some of its review of information retrieval systems. However, it offers much new content in the post-genomic era of informatics. The work is most clearly appropriate for readers in medical informatics and bioinformatics who are focused on building applications, and performing research, in the rich and complex information environment that the genome mapping projects of the last decade have provided us. This is not to say that Shortliffe has not incorporated some of this material, and reoriented the focus more broadly, as appropriate. Shortliffe’s third edition adds several completely new chapters on “Natural Language and Text Processing in Biomedicine” and “Imaging and Structural Informatics,” and a substantially rewritten chapter on “Imaging Systems in Radiology.”
At this point, Chen and Shortliffe’s works remain complementary texts, suitable for introductory graduate-level courses in medical informatics and bioinformatics, and with wide uses in a variety of other disciplines, as noted above. Shortliffe provides a broader introduction to the field, whereas Chen is most valuable for more specific data mining and knowledge management applications.