Session S74.1

The PhysioNet / Computers in Cardiology Challenge 2008: T-Wave Alternans

GB Moody*

Massachusetts Institute of Technology
Cambridge, MA, USA

One hundred years after it was first reported by Hering, T-wave alternans (TWA) is widely understood to be an important indicator of risk of sudden cardiac death. Yet for most of that time TWA was believed to be rare, until at this meeting in 1981, when Adam, Akselrod, and Cohen reported the existence of microvolt-level TWA, too small in amplitude to be detected visually at standard display scales. Many algorithms for detecting and quantifying TWA have been proposed, but it remains very difficult to validate or to compare any of these algorithms since no generally accepted objective criteria exist for measuring TWA, and no generally available set of validation data exists as a basis for comparison. This Challenge aims to improve understanding of the strengths and limitations of classic and novel TWA analysis methods; to establish a collection of reference ECGs ranked by severity of TWA, as determined by a preponderance of evidence; and to encourage the development and dissemination of open-source TWA detectors and estimators in order to support and stimulate further research on the properties and implications of TWA. Interested researchers were invited to nominate ECG recordings for inclusion in the Challenge data set. On the basis of the nominations, 100 two-minute segments were posted on PhysioNet, including ECGs from subjects with a variety of risk factors for sudden cardiac death (acute myocardial infarctions, transient ST depression, and ventricular fibrillation) as well as healthy controls and synthetic TWA in calibrated amounts. Challenge participants analyzed these segments, and submitted measurements of TWA in each segment for scoring. Since the various analyses produce incommensurate measurements, each participant's measurements were sorted to provide a list of the 100 records in order of estimated TWA severity. These lists were aggregated to derive a reference ordering, then the agreement between each participant and the reference ordering was evaluated (by calculating the Kendall rank correlation coefficient). As of 30 April, 12 entries had been submitted, achieving preliminary scores between 0.4 and 0.8. Challenge participants will continue to submit entries until 1 September, when final scores will be determined. At the conclusion of the Challenge, the collected TWA measurements will be made available for further study.

(Abstract Control Number: 384)