The Use of Artificial Intelligence Systems for Support of Medical Decision-Making

The Use of Artificial Intelligence Systems for Support of Medical Decision-Making

William Claster (Ritsumeikan Asia Pacific University, Japan), Nader Ghotbi (Ritsumeikan Asia Pacific University, Japan) and Subana Shanmuganathan (Auckland University of Technology, New Zealand)
DOI: 10.4018/978-1-60566-266-4.ch002
OnDemand PDF Download:
No Current Special Offers


There is a treasure trove of hidden information in the textual and narrative data of medical records that can be deciphered by text-mining techniques. The information provided by these methods can provide a basis for medical artificial intelligence and help support or improve clinical decision making by medical doctors. In this paper we extend previous work in an effort to extract meaningful information from free text medical records. We discuss a methodology for the analysis of medical records using some statistical analysis and the Kohonen Self-Organizing Map (SOM). The medical data derive from about 700 pediatric patients’ radiology department records where CT (Computed Tomography) scanning was used as part of a diagnostic exploration. The patients underwent CT scanning (single and multiple) throughout a one-year period in 2004 at the Nagasaki University Medical Hospital. Our approach led to a model based on SOM clusters and statistical analysis which may suggest a strategy for limiting CT scan requests. This is important because radiation at levels ordinarily used for CT scanning may pose significant health risks especially to children.
Chapter Preview


The advent of computed tomography (CT) has revolutionized diagnostic radiology (Figure 1). Since the inception of CT in the 1970s, its use has increased rapidly. It is estimated that more than 62 million CT scans per year are currently obtained in the United States, including at least 4 million for children (Brenner & Hall, 2007). The increase in the use of medical radiation, especially in diagnostic CT scanning has raised many concerns over the possible adverse effects of procedures conducted in the absence of any serious risk/benefit analysis, especially where these procedures are carried out on

Figure 1.

The basics of CT


Children (UNSCEAR, 2000). According to a survey conducted in 1996 (White, 1996) the number of CT scanners per 1 million population was 26 in the United States and 64 in Japan. The growth of CT use in children has been driven primarily by the decrease in the time needed to perform a scan — now less than 1 second — largely eliminating the need for anesthesia to prevent the child from moving during image acquisition (Frush et al, 2003). Overuse of diagnostic CT radiation, which deliver radiation doses 50 to 200 times higher than most X-rays, can lead to an increased risk of cancer. Additionally, it may lead to an unnecessary rise in health care costs (Roebuck, 1999; Ghotbi et al, 2005; Walsh, 2004).

Complete Chapter List

Search this Book: