NASA / JDF

Abstract of the talk/presentation at the JDF/NASA Non/Minimally Invasive Measurement of Biological Analytes Technology Workshop (April 8-9, 1998 - Washington D.C.)

Integrated Optoelectronic Sensor Suitable for Multivariate Analysis of Complex Spectra

Aureliu M. Porumbescu
Array Vision Engineering Co.

Array Vision Engineering’s (AVE’s) approach to noninvasive analyte determination combines an optoelectronic sensor with an individual-specific, knowledge-based metabolic modeling and prediction methodology. Blending computerized kinetic modeling with a suitable position-sensitive photodetector will improve the efficiency of multivariate analysis of complex spectra obtained from living subjects in real time.

Managing diabetes involves an individual about whom considerable knowledge exists, and knowing when to test blood glucose is usually more important than the test results. With careful consideration of all major inputs and outputs, models can predict much of what is occurring; therefore, individual-specific modeling is as important as the measurement method. Rather than aiming for measurement accuracy and precision, AVE’s objective is to provide the individual with sufficient information to achieve proactive metabolic control, including the optimal timing and specifics of key activities such as insulin injections, blood glucose measurements, meals, exercise, and so forth. AVE’s method to improve metabolic control is to determine a patient-specific profile and then predict individual outcomes on an ongoing basis, giving the user information that facilitates corrective action before an out-of-control condition occurs.

For the optoelectronic sensor, AVE developed a wedge-and-strip staring-array detector capable of computing the coordinates of the centroid of a spot of light incident on its surface. Unlike customary charge coupled devices, this detector uses simple analog electronics and an algorithm with reduced sensitivity to electronic and partition noise, yielding real time data at video rates. Massive parallelism and multiobject detection capability with spatial prefiltering, may make this the optimal solution for many applications involving low level or gated light signals.

Decreasing costs for computers and communications and increasing pressures to contain health care costs promote technology-intensive proactive approaches to health care, especially in diabetes, but AVE’s methodology is also applicable to any other medical condition where measurements are routinely made. The main benefit is that the patient has the ability to implement proactive actions that do not drastically affect his or her lifestyle, which leads to greatly improved health outcomes and compliance.