Use of big data in medicine

Abstract

As a result of the advances in computers, data collection has become easy. The huge amount of data now available through computers is called big data. The characteristics of big data are represented by three Vs: (1) volume, (2) variety, and (3) velocity. Volume, variety, and velocity refer to the amount of data, the range of data types and sources, and the speed and frequency of the acquisition of responses from users, respectively. The steps in the analysis of big data include (1) the review of data, (2) the construction of a model, and (3) the interpretation of results. Big data has mainly been analyzed by methods based on correlation analysis. The analytical method mainly based on the correlations found in big data is useful in predicting the prognosis from the analytical results. However, the factors useful for predicting the prognosis are not always useful in the treatment of patients. In medicine, the results of correlating events alone cannot be applied to clinical practice because the causal relationship between causes and the onset of diseases is important. To use big data in medicine, the results obtained from the analysis of big data should be validated by clinical studies.