Impedance Spectroscopy for Biofouling Detection

Electrochemical impedance spectroscopy (EIS) sends a small alternating voltage into an electrode at a series of frequencies and records how the current answers back [1]. Researchers have asked whether that answer could reveal biofouling, the film of proteins and other biological material that coats an implanted glucose sensor and slows the glucose reaching it [2]. As of October 2026, we found impedance-based sensor checks in lab studies and in a Medtronic MiniMed patent filing [3], but not in the FDA decision summaries we read for current CGMs [4][5][6][7]. So we can't tell you that any sensor on the market runs them.

What biofouling does to a sensor

Biofouling begins in the first day after insertion [8]. A CGM's sensing electrode sits under a membrane that lets glucose diffuse in, and proteins from the surrounding fluid stick to that membrane, limit how much glucose gets through, and lower the sensor's sensitivity [2]. A University of Kansas team implanted needle-type glucose sensors in rats for 24 hours and found mostly fragments of serum albumin on and inside the membranes. They concluded those fragments cut glucose diffusivity, and their paper cites biofouling as the primary reason sensors lose sensitivity in the first 24 hours after implant [8].

Here's what surprised us: the coating on the device wasn't the biggest barrier in one of the field's key studies. Duke researchers measured glucose flow through membranes on probes implanted in rats over 8 days. The fouling layer stuck to the probe accounted for only 12 to 24% of the resistance to glucose transport, and the tissue around the probe accounted for far more [9].

That changes what "biofouling detection" really has to detect. The target is the whole boundary between sensor and tissue, not just a film on the membrane. It also fits a long-running frustration in this field, which a 2007 review put plainly: it remains hard to predict whether an implanted glucose sensor will work reliably, slowly lose performance, or fail soon after implantation [10]. We cover the body's side of this in Sensor Bio-fouling and Biocompatibility.

How impedance spectroscopy works

Impedance is resistance that depends on frequency. In an EIS measurement, an instrument lays a small alternating voltage over a steady one and reads the alternating current that comes back [1]. Fast processes at the electrode show up at high frequencies. Slow ones show up at low frequencies, so sweeping across many frequencies pulls apart effects that would otherwise blur together [1].

The tutorial we relied on works through an example. Charging of the double layer (the thin sheet of electric charge where metal meets liquid) appears near 4.1 kHz, the electrode reaction near 310 Hz, and diffusion of molecules toward the surface near 0.4 Hz [1]. That's the reasoning behind proposals to use EIS on implanted sensors: biofouling works by slowing diffusion [2], and diffusion sits at the slow end of the spectrum [1].

The catch is time. A single measurement at 1 millihertz takes about 17 minutes, and a sweep from 100 kHz down to 0.1 Hz takes 2 to 3 minutes [1]. A wearable sensor would have to fit those sweeps around its main job of reading glucose. We found no published figure for what that would cost in battery life on a CGM transmitter.

What studies of implanted electrodes show

The clearest evidence that impedance can track tissue around an implant comes from brain electrodes, not glucose sensors. Researchers at Georgia Tech and Emory recorded impedance spectra from microelectrodes implanted in rats and compared them with tissue stains afterward [11]. Several impedance measures correlated with a marker of astrocyte scarring called GFAP, and the correlations were confined to the first 100 micrometers around the electrode [11]. Scar tissue in the brain isn't the capsule that forms under the skin, so we read this as a proof of concept for implanted electrodes in general.

Work on glucose sensors themselves is thinner. A 2018 IEEE Sensors Letters paper by Ghoreishizadeh and colleagues is titled "Study of Electrochemical Impedance of a Continuous Glucose Monitoring Sensor and its Correlation With Sensor Performance" [12]. Its title names exactly the link this approach depends on.

Keep one distinction in mind when you read about this topic. Some groups use impedance to measure glucose itself rather than to check on a glucose sensor. A Swedish team built a battery-less implantable bioimpedance spectroscope and tested it in pigs; across 236 measurements under the skin during glucose tolerance tests, its estimates correlated very well with blood glucose reference values [13]. That's a way of sensing glucose, not a biofouling test.

What Medtronic MiniMed patented

The industry document we found is a patent application. US 2011/0230741 A1, "Methods and Systems for Observing Sensor Parameters," names Medtronic MiniMed, Inc. as assignee and was published September 22, 2011, as a division of an application filed in December 2008 [3]. It describes applying a voltage, such as a voltage pulse, to a glucose sensor and deriving values like impedance and capacitance from the current response [3]. The filing treats those values as signs of the sensor's state, including hydration, noise, offset and drift, and one of its figures is labeled "24-hour EIS sampling" [3].

Its focus is the start of wear, not biofouling. The background section says that with many continuous glucose sensors, the user had to wait three hours for the sensor to stabilize before its readings were used [3]. A patent shows what Medtronic claimed, not what any MiniMed sensor does. For how to read filings like this one, see Patent Landscape and Intellectual Property.

Does any marketed CGM use it?

We couldn't confirm one. As of October 2026, we searched the full text of four FDA decision summaries: the Dexcom G7 clearance (K213919) [4], the clearance that added the FreeStyle Libre 3 Plus and Libre 2 Plus sensors (K233537) [5], the MiniMed 780G approval that added the Guardian 4 sensor (P160017/S091) [6], and the 2025 supplement that added the Simplera Sync sensor (P160017/S118) [7]. None of them contains the word "impedance."

Silence in a summary isn't proof. These documents focus on intended use and clinical performance [4][6], and they don't list every check a transmitter's firmware runs. What we can say is narrower: no FDA summary we read tells users that their sensor detects biofouling by measuring impedance.

What's still unknown

The hardest part is timing. The first hours after insertion are when a sensor is still settling: the Medtronic filing describes a multi-hour stabilization wait [3], and the Kansas group ties early sensitivity loss to biofouling within the first 24 hours [8]. An impedance check in that window would have to tell a sensor that is still hydrating from one that is already fouling. We haven't found a published study showing a wearable CGM doing that.