Application of Data Mining Techniques in Clinical Decision Making: A Literature Review and Classification

Application of Data Mining Techniques in Clinical Decision Making: A Literature Review and Classification

Hakimeh Ameri (K. N. Toosi University of Technology, Iran), Somayeh Alizadeh (K. N. Toosi University of Technology, Iran) and Elham Akhond Zadeh Noughabi (University of Calgary, Canada)
DOI: 10.4018/978-1-5225-2515-8.ch012
OnDemand PDF Download:
$37.50

Abstract

Data mining techniques are increasingly used in clinical decision making and help the physicians to make more accurate and effective decisions. In this chapter, a classification of data mining applications in clinical decision making is presented through a systematic review. The applications of data mining techniques in clinical decision making are divided into two main categories: diagnosis and treatment. Early prediction of medical conditions, detecting multi-morbidity and complications of diseases, identifying and predicting the chronic diseases and medical imaging are the subcategories which are defined in the diagnosis part. The Treatment category is composed of treatment effectiveness and predicting the average length of stay in hospital. The majority of the reviewed articles are related to diagnosis and there is only one article which discusses the determination of drug dosage in successful treatment. The classification model is the most commonly practical model in the clinical decision making.
Chapter Preview
Top

Research Methodology

The following online journal databases were searched to provide a bibliography of the clinical decision making and data mining.

  • Science direct

  • IEEE Transaction

  • Hindawi Publishing

  • Pub med

  • Springer

  • Weily online library

  • Google scholar

  • Online international conferences

Approximately 600 articles were read and reviewed. Those articles were not completely related to the subjects of applied data mining in clinical decision making were eliminated at the first step. Some conference papers with a low citations, text- books, masters and doctoral dissertations, and unpublished working papers were eliminated.

After this elimination we reached to about 200 papers for our research. Each article was reviewed again and two main categories and nine subcategories of clinical decision making were recognized. These categories and sub categories provide a comprehensive base for understanding of data mining research for clinical decision making.

Top

Classification Method

As mentioned earlier, we categorized the applications of data mining techniques in clinical decision making into two main categories and nine sub-categories. These two categories include diagnosis and treatment that each one has some sub-categories. On the data mining side, six modeling types including classification, clustering, feature extraction, association analysis, regression and time series were recognized. We explain the categories and different types of modeling in the following in more details.

Key Terms in this Chapter

Diagnosis: Diagnosis is the act of the identification of a disease, illness, or a problem by the examination of something or someone’s inspection. The diagnosis is based on information from some sources such as results of a physical examination, interview with the patient, family, or both, the medical history of the patient and family, and clinical findings that reported by laboratory tests and radiologic studies.

Data Mining: Data mining is the computing process of discovering patterns in large data sets involving methods at the intersection of artificial intelligence, machine learning, statistics, and database systems. The overall goal of the data mining process is to extract information from a data set and transform it into an understandable structure for further use.

Treatment: Treatment is defined as the management and medical care of a patient to combat disease or disorder.

Complete Chapter List

Search this Book:
Reset