Optical spectroscopy can be used to determine the concentration of chemical species in samples. The amount of light absorbed by a particular chemical species is often linearly related to its concentration through Beer’s Law. For nontransparent materials such as powders, tablets, natural materials (soil), blood, skin, and muscle, optical information can be collected via diffuse reflectance spectroscopy.
Diffuse reflectance spectroscopy techniques with near infrared spectroscopy (NIRS) have been used for the noninvasive measurement of blood and tissue chemistry in human and animal subjects. NIRS can provide accurate and continuous measurement of medical analytes without the need to remove a blood or tissue sample from the patient. The application of this technique involves shining near-infrared light onto the skin directly or through a fiber optic bundle, and measuring the spectrum of the light that is reflected back from the blood-containing muscle. When diffuse reflectance NIRS is used to measure blood hematocrit in muscle or organs, the accuracy of the measurements can be affected by absorbance variations in layers overlying the muscle or organs (e.g., due to variations in the thickness of fat and skin layers between different patients in a patient population, or between different locations on an individual patient).
Sophisticated mathematical techniques, such as partial least squares regression and other multivariate calibration methods, are used to determine a relationship between concentration and absorbance. Once these calibration models are derived, they can be used to determine chemical composition by measuring absorbance in transmittance or reflectance mode.
Chemometrics is a branch of chemistry that provides statistics-based techniques to process multi-wavelength spectra such that analytic concentration can be calculated from the reflectance spectra recorded from complex media such as biological tissue. Chemometrics is used to derive a mathematical relationship between relevant portions of the spectra collected from a sample and the concentration or amount of the analyte of interest in the sample. The relationship between the spectra and the chemical concentration can be expressed as a “calibration equation” that can be programmed into a patient monitor and used to determine analytic concentrations based on the measured reflectance spectra. Spectra collected from patients can be processed through calibration equation(s) stored in the patient monitor, and the analyte concentration in those patients can be reported based on the collected spectra and the calibration equations.
These measurement systems and methods include a light source and a detection system. They also include a set of at least first, second, and third light ports that transmit light from the light source to a sample and receive the direct light reflected from the sample to the detection system. This generates a first set of data including information corresponding to both an internal target within the sample, and features overlying the internal target, as well as a second set of data including information corresponding to features overlying the internal target. A processor is configured to remove information characteristic of the overlying features from the first set of data using the first and second sets of data to produce corrected information representing the internal target.