Computers have proved people wrong again. What were pushed aside as "noises" in ECG data have now been identified--thanks to researchers' use of data-mining and machine learning methods--as biomarkers for an increased risk of death after certain cases of acute coronary syndrome.
Using existing ECG data from 4,557 heart attack patients who took part in a clinical trial, researchers from the University of Michigan, Ann Arbor used the computational methods to identify three previously undiscovered biomarkers for higher risk of cardiovascular death, MedPage Today reported. The findings were recently published online in Science Translational Medicine. The three biomarkers are all patterns found in the ECG data.
"The main point of our study is that we are able to make better use of data that we're already collecting," Zeeshan Syed, of UMich, told MedPage Today. "There's prognostic information buried in the noise, and it's almost invisible because of the sheer volume of the data. The sophisticated computational techniques allow us to home in on truly abnormal ECG signals. These patients with unstable hearts are at a greater risk of dying. Identifying them would allow physicians to initiate more aggressive treatment."
With retrospective analysis done, the group wants to test the biomarkers in patients in the hospital. Syed told the publication that he believes the biomarkers could be used alongside existing methods of categorizing high-risk patients with ACS.