Data mining techniques for stroke: a systematic review

Abstract

The objective of this study was to conduct a systematic review of applications of data mining techniques in the field of stroke research. We searched the MEDLINE database through PubMed and PubMed Central. We initially identified 28 articles (17 from PubMed and 11 from PubMed Central) by the search, and selected 13 articles representing various data mining methods used for stroke research. Our main interest was to identify research goals, stroke types, datasets, and data mining methods, data mining software and outcomes. The applications of data mining techniques in the selected articles were useful for extracting valuable knowledge and generating new hypothesis for further research work, prediction of stroke's risk factor and improving healthcare for stroke patients. The results could be used for scientific research and real-life practice to improve the quality of healthcare stroke patients. Data mining has played an important role in stroke research. Data mining would be a valuable asset for stroke researchers because it can unearth hidden knowledge from a huge amount of stroke-related data. We believe that data mining can significantly help stroke research and ultimately improve the quality of healthcare for stroke patients.