Blood Protein Predicts CAD with 94% Accuracy
Bio-MedPoint’s predictive tool uses gel capillary electrophoresis to detect the protein fingerprint in the patient’s blood. The electric charge in the blood serum medium causes the proteins to travel at different speeds through the molecule mixture.
As the proteins pass through the capillary window a detector records the light transmitted through an electropherogram. The UV absorbance pattern uses 32 Karat Software to analyze the data and create a graph with peaks and valleys.
Simple Blood Test
Bio-MedPoint’s screening tool uses a small sample of blood serum— either from the vein or a finger-stick— and passes it through a quick screening tool.
Computer algorithms not only show evidence of the disease, they also indicate varying degrees of disease— from healthy to high probability of CAD. Then it lists the results in a Linear Discriminant Analysis (LDA) score.
Bio-MedPoint’s algorithms allow it to distinguish between healthy protein rates and spikes in biomarkers that indicate Coronary Artery Disease (CAD) disease. This system uses the electrophoresis in a unique way in that it separates the biomarkers by mass not charge.
Known molecular weights are displayed on the electropherogram axes, which presents the migration time (Minutes) versus absorbance units (AU). Weights of macromolecules can be recorded based on migration time thus identifying biomarkers.
Univariate analysis of variance (ANOVA) helps identify biomarkers involved either directly or indirectly in health and disease. They do this by comparing the CGE profiles from the biofluid of diseased and/or healthy individuals.
The peaks represent macromolecules that may be contributors to CAD classifications. These may be markers of the disease or Non-CAD classifications and indicators of arterial health. Then the system computes LDA scores for each participant. A negative score indicates a low probability of CAD and a positive score indicates a high probability of CAD.
When testing the system, professors at a major university in the U.S. accurately classified 94% of the study participants as having or not having coronary artery disease based on serum profiles. This compared to 42% correct with the FRS classification.