Computational Prediction Models for Early Detection of Risk of Cardiovascular Events Using Mass Spectrometry Data

• Published in: IEEE transactions on information technology in biomedicine Vol. 12; no. 5; pp. 636 - 643
• Main Authors: Pham, T.D., Honghui Wang, Xiaobo Zhou, Dominik Beck, Brandl, M., Hoehn, G., Azok, J., Brennan, M.-L., Hazen, S.L., Li, K., Wong, S.T.C.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.09.2008
• Subjects: Algorithms; Biomarkers - blood; Blood; Blood Chemical Analysis - methods; Blood Proteins - analysis; Cardiology; Cardiovascular Diseases - blood; Cardiovascular Diseases - diagnosis; Cardiovascular risk; Computational modeling; Data mining; Diagnosis, Computer-Assisted - methods; early disease detection; Event detection; Humans; mass spectrometry (MS); Mass Spectrometry - methods; Mass spectroscopy; Medical diagnostic imaging; Pain; prediction models; Predictive models; proteomics; Reproducibility of Results; Risk Assessment - methods; Risk Factors; Robustness; Sensitivity and Specificity; cardiovascular risk; Prediction models; Early Disease detection; Mass spectrometry; Proteomics
• ISSN: 1089-7771; 1558-0032
• DOI: 10.1109/TITB.2007.908756

Early prediction of the risk of cardiovascular events in patients with chest pain is critical in order to provide appropriate medical care for those with positive diagnosis. This paper introduces a computational methodology for predicting such events in the context of robust computerized classification using mass spectrometry data of blood samples collected from patients in emergency departments. We applied the computational theories of statistical and geostatistical linear prediction models to extract effective features of the mass spectra and a simple decision logic to classify disease and control samples for the purpose of early detection. While the statistical and geostatistical techniques provide better results than those obtained from some other methods, the geostatistical approach yields superior results in terms of sensitivity and specificity in various designs of the data set for validation, training, and testing. The proposed computational strategies are very promising for predicting major adverse cardiac events within six months.