Cloud-based platform may make ECG data easier to assess

A new Web-based platform that analyzes multimodal signals may help make electrophysiological data easier to evaluate and incorporate into clinical research, according to a study published in the March issue of the Journal of the American Medical Informatics Association.

Researchers from Case Western Reserve University in Cleveland used the Cloudwave platform and assessed its potential for facilitating the use of "big data" such as electrophysiological data with many signals. They compared its data evaluations to compute cardiac measures, such as QRS complexes, with the evaluations of traditional desktop methods on data from patients with epilepsy.

The epilepsy data had been gathered at epilepsy monitoring units at the university as part of the National Institute of Neurological Diseases and Stroke’s Prevention and Risk Identification of SUDEP [sudden unexplained death in epilepsy] Mortality (PRISM) project. The PRISM project currently uses Cloudwave.

The investigators, led by Satya S. Sahoo, PhD, found that computations took less time using Cloudwave when compared with the desktop approaches. Cloudwave reduced the time by a factor of 3.8 (0.32 vs. 1.2 minutes) for data from one electrocardiogram (ECG) channel. The platform was one order of magnitude faster (4.8 vs. 0.48 min) than the desktop computer for data from four ECG channels.

They added that Cloudwave allows for the secure control of access to data, complying with the Health Insurance Portability and Accountability Act. They also noted that Cloudwave usage cost about 24 cents per hour for large instances and 48 cents per hour for extra large instances, which was comparable to commercial cloud infrastructures. Storage cost was 10 cents per gigabyte for one month, also comparable to commercial platforms. However, commercial infrastructures could add the additional cost of data transfer.

In conclusion, explained the authors, Cloudwave was able to store a large volume of signal data, perform complex computations and support ontology-driven web-based visualization and query access used in collaborative research efforts.