Senseonics is attempting to develop a glucose sensor that would be implanted under the skin for a full year. Their novel approach uses an extremely tiny light-emitting diode or LED near a separate photodiode receiver, like the ones in solar powered calculators, that was able to measure light.
All of this is encapsulated into a pill-shaped structure about the size of a medicine capsule.
What makes the device really work is actually the new part. Both are widely available and cost pennies apiece. Their small size and low cost open the door to new applications, like tiny sensors that can be implanted under the skin.
Now SMSI would be . Between the two would be the ambient ISF glucose and some theoretical molecule that fluoresces in the presence of glucose. The more glucose, the more (or less) fluorescence which is then measured by the photodiode with a signal transmitted to the outside.
Large hurdles remain, especially the discovery of a fluorescent molecule that will fluoresce over the desired measurement range for glucose, stabilization of the fluorescent particle within the hardware, development of miniturized and reliable hardware, and transmission of the measurements outside the body without a significant increase in size or power requirements. There are also many questions, like what metal or plastic materials will be involved and will they be simply left under the skin or removed when no longer working. This concept will float or sink on whether a non-toxic fluorescing molecule can be found that would work over the desired glucose range. But the approach remains interesting.
Implantable Glucose Monitor
A microchip being developed by Sensors for Medicine and Science Inc. (SMSI; Germantown, MD) incorporates an optical sensor capable of detecting a wide range of analytes, including oxygen, carbon dioxide, pH, glucose, lactate, and anesthetic gases. The firm's oxygen sensor prototypes use a tiny light-emitting diode (LED) as the light source and an ordinary photodiode to measure the light output. Both low-cost components are widely available. A key element of the SMSI design is the placement of the excitation source in the sensor element. The LED is embedded in a matrix containing the fluorescent indicator molecules to overcome a number of technical challenges. "If you understand how light behaves in the presence of certain compounds, you can use light to detect the presence or absence of those compounds," says Arthur Colvin, who designed the sensor platform. "The hard part is using what you know to build a sensor that's stable, very sensitive, and inexpensive."
The small size and low cost of the SMSI sensor components are essential to new applications, including implantable microsensors, according to the company. Conventional glucose-monitoring systems require skin pricks and blood sampling. SMSI is now working on a sensor small enough to be injected into the fatty layer below the skin to measure blood sugar levels in diabetic patients. Sugar reduces the light reemitted by the fluorescent chemical. Basically, the higher the glucose level, the less light is detected. The implanted SMSI sensor would thus measure the glucose level by detecting the reduced light level. The result would be transmitted to an external unit via radio-frequency communication. The low-power unit, resting just beneath the surface of the skin, would be powered externally by induction.
The company indicates that the optimal fluorescent chemical remains in development; however, the chip design has been completed. The prototype device uses a 22-µW LED--a fraction of the power needed for the power-on indicators of most personal computers. Because the fluorescence detection process consumes no chemicals or proteins, the device is self-sustaining, according to the firm.
"Diabetes is a disease that typically attacks people when they are very young or as they begin to grow old," SMSI president and COO Marc Schneebaum notes. "Those are the times in people's lives when they may be most vulnerable. We think that's a very compelling reason for the work we're doing." Schneebaum adds that "an implantable sensor that could be powered up and queried on demand would go a long way toward helping diabetics manage the disease."