Support Vector Machine for Hematocrit Density Estimation Based on Changing Patterns of Transduced Anodic Current

In measuring the glucose density in the whole blood by handheld devices, the hematocrit is one of the most highly influencing factors. Since the hematocrit dependency should be appropriately controlled for accurate measurement of glucose density, information regarding hematocrit should be obtained at reasonable cost. This paper introduces an approach for hematocrit estimation from the transduced anodic current curves using support vector machine. These current curves are produced by glucose-oxidase reaction in the strip-type electrochemical biosensors which is commercially available. The support vector machine used in this paper was trained for regression with the target value of accurate glucose values measured by a hospital analysis system. This method can obtain an acceptable result that can be used to reduce the dependency of hematocrit in the measurement of glucose values in the whole blood with hand-held meters.