The developing world does not have access to many of the best medical diagnostic technologies; they were designed for air-conditioned laboratories, refrigerated storage of chemicals, a constant supply of calibrators and reagents, stable electrical power, highly trained personnel and rapid transportation of samples. Microfluidic systems allow miniaturization and integration of complex functions, which could move sophisticated diagnostic tools out of the developed-world laboratory. These systems must be inexpensive, but also accurate, reliable, rugged and well suited to the medical and social contexts of the developing world.

Biomedical engineers have traditionally developed technologies in response to the needs of the developed world's medical community. As a result, the diagnostic systems on which they have worked have met the requirements of well-funded laboratories in highly regulated and quality-assessed environments. However, such approaches do not address the needs of the majority of the world's people afflicted with infectious diseases, who have, at best, access to poorly resourced health care facilities with almost no supporting clinical laboratory infrastructure. A major challenge for the biomedical engineering community is to develop diagnostic tests to meet the needs of these people, the majority of whom are in the developing world. We here review the context in which the diagnostics must operate, some of the appropriate diagnostic technologies already in distribution, and some emerging technologies that promise to address this challenge. However, there is much room for innovation, adaptation, and cost reduction before these technologies can impact health care in the developing world.

The T-sensor is a recently developed microfluidic chemical measurement device that exploits the low Reynolds number flow conditions in microfabricated channels. The interdiffusion and resulting chemical interaction of components from two or more input fluid streams can be monitored optically, allowing measurement of analyte concentrations on a continuous basis. In a simple form of T-sensor, the concentration of a target analyte is determined by measuring fluorescence intensity in a region where the analyte and a fluorescent indicator have interdiffused. An analytical model has been developed that predicts device behavior from the diffusion coefficients of the analyte, indicator, and analyte--indicator complex and from the kinetics of the complex formation. Diffusion coefficients depend on the local viscosity which, in turn, depends on local concentrations of all analytes. These relationships, as well as reaction equilibria, are often unknown. A rapid method for determining these unknown parameters by interpreting T-sensor experiments through the model is presented.

at room temperature in an aqueous solution, a spherical molecule with a molecu-MicrofLuidicDiffusion-Based larweightof330(forexample,asmallorganic dye molecule) takes 0.2 s to diffuse Separation and DetectionTO pm, whereas a larger particle with a diameter of 0.5 pm (for example, a small bacterium) would require about 200 s to