Identification and compression ratios of standards for recording time-series resulting from biomedical signals

Abstract

Over the years, the development of technologies in the healthcare sector has been fundamental for the identification of diseases, time reduction of clinical assessments and patient monitoring. The main implication of this is an increase in the quality of care provided to patients. However, the use of technology has some relevant side effects, being the increase in the amount of produced information one of the most important. Currently, there is a large volume of medical devices able to collect vital signs from patients, and this diversity generated by different equipment manufacturers, originates a variety of file formats. This factor complicates the sharing of data between research centers. In this context, this study identified the main file formats existing for the storage of time-series resulting from biomedical signals. The distinct file formats were assessed by using a benchmark, which was biomedical data publicly available. It was found that the file format EDF is the one which requires less memory storage capacity and the possibility of accounting for embedding structured information relevant for the description of the data (e.g., annotation of events, patient and device information). Considering the increase in the amount of digital information and the eminent necessity of sharing data among researchers these results suggest that researchers should adopt available solutions, for instance the use of the EDF file format, for dealing with the problem of data compression and dissemination.