Serum protein profiles predict coronary artery disease in symptomatic patients referred for coronary angiography
1 University of Pittsburgh, Department of Pathology, 5230 Centre Avenue, Pittsburgh, PA 15232, USA
2 University of Pittsburgh, Department of Biomedical Informatics, 5150 Centre Avenue, Pittsburgh, PA 15232, USA
3 Materials Processing Center, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
4 Prevencio LLC, 454 N 34th Street, Seattle, WA 98103, USA
5 University of Pittsburgh, Division of Cardiology, Department of Medicine, 200 Lothrop Street, Pittsburgh, PA 15213, USA
6 UPMC Heart and Vascular Institute, 200 Lothrop Street, Pittsburgh, PA 15213, USA
7 University of Pittsburgh, Department of Epidemiology, Graduate School of Public Health, Pittsburgh PA, 15213, USA
BMC Medicine 2012, 10:157 doi:10.1186/1741-7015-10-157Published: 5 December 2012
Additional file 1:
Algorithm development. The analytical form of the scoring function is provided. The testing that was performed to derive and optimize the algorithm is described including the use of artificial markers and cross validation testing.
Format: DOCX Size: 25KB Download file
Additional file 2:
Predictive panel selection. The process of testing and final derivation of the panels comprising two, three, four or five markers that best predicted the classification outcome of the patients based on coronary angiography is provided.
Format: DOCX Size: 24KB Download file