Low-cost near-infrared measurement of subcutaneous fat for newborn malnutrition

摘要

Low fat composition in newborns exposes them to an immediate risk of increased mortality and morbidity, inhibited physical and cognitive development and to diabetes and obesity diseases in later life. Information about nutritional and dietary status of newborns can be accessed by measuring the amount of fat composition in the body. The functions of subcutaneous fat involve energy storage, thermo-insulation and a physical buffer. Current technologies for newborn body fat monitoring are: a device based on air displacement plethesmography (PeaPod), dual-energy X-ray, and underwater weighting. However they are bulky, expensive, immobile, and require technical expertise. We propose an alternative portable measurement system of in-vitro for subcutaneous fat that uses diffuse near-infrared light reflectance measurement system. We also introduce an in-vitro three-layered tissue model mimicking the subcutaneous fat layer in newborns together with a preliminary study to measure fat using dual-wavelength near-infrared light. Based on the output data from these measurements, we have proposed a suitable transmission and scattering model. This model estimated the amount of reflected light collected by a photodetector after incident light is scattered in several fat layers. Our portable sensor is low cost and does not require training hence it is suitable for mass use in the developing world. It consists of a single LED and two photodetectors (900 nm and 1000 nm). The photodetectors wavelengths were chosen to be sensitive to fat as it exhibits a peak in the wavelength at 930 nm and to water at which exhibits a peak at 980 nm; the latter is used, to remove hydration bias. Results on a porcine tissue model demonstrate differentiation as low as 2 mm fat which is a relevant screening thickness to indicate low percentage body fat.