A certain band of electromagnetic radiations is used in optical technologies for testing with Non-Invasive Blood Glucose Monitoring Devices. These radiations interact with tissues' constituents, including glucose, after spreading through the tissue. The spectrum, which was gathered during light propagation, is used to determine the concentration of glucose in the sampled tissue volume. Selectivity is one of the most crucial parameters in non-invasive glucose sensing. Selectivity is the ability to identify a specific concentration of an analyte in a complicated matrix without the interference of other substances. The response is based on the interactions that are mostly accessed by multivariate data analysis, and a number of selectivity-generating stages (detection and multistage separation principles) are usually utilized to overcome the influence of interfering interactions (chemometrics).
Since there are many metabolites present in the human body that have an optical signature similar to glucose 30, sensor selectivity is a crucial issue in optical methods of glucose detection. A proper mathematical model should be incorporated for reaching the selectivity. Selectivity problems have been investigated in several multivariate calibration models, which has hindered the development of Non-Invasive Blood Glucose Monitoring Devices sensing with various approaches. Nevertheless, the addition of more measurements can enhance selectivity.
