GB2584420A

GB2584420A - Method, sensor and system for determining a dielectric property of a sample

Info

Publication number: GB2584420A
Application number: GB1907348.5A
Authority: GB
Inventors: Kale Izzet; Turgul Volkan
Original assignee: University of Westminster
Current assignee: University of Westminster
Priority date: 2019-05-24
Filing date: 2019-05-24
Publication date: 2020-12-09
Also published as: GB201907348D0; WO2020240157A1

Abstract

Method and system for determining a dielectric property of a sample, optionally a human finger for blood-glucose concentration monitoring. The method uses a system comprising an oscillator circuit 10 including a device, which may be a radio frequency (RF) sensor 12, for physically coupling to a sample, measuring at least one frequency generated by the oscillator, and calculating a dielectric property based on the measured frequency. Two separate frequency signals may be generated (optionally one below 100MHz and one above 500MHz) where the shift in the lower frequency indicates a degree of coupling between sample and device, allowing calculation of a calibration factor, and the second is indicative of a dielectric property of the sample, allowing calculation of a dielectric property using the calibration factor and second frequency shift. The sample may be plant or animal tissue. An independent claim is made to a system comprising a processor for performing measurements of frequency and calculating dielectric properties, optionally including a phase-locked-loop or spectral analyser.

Description

METHOD, SENSOR AND SYSTEM FOR DETERMINING A DIELECTRIC PROPERTY OF

A SAMPLE

FIELD OF THE INVENTION

The present invention relates to a method, sensor and system for determining a dielectric property of a sample.

RELATED ART

Blood glucose concentration is an important biometric for diagnosing and monitoring diseases such as diabetes. A common method for measuring blood glucose levels is via a finger-prick test or in a lab setting through various measurements including an HbAlc test, whereby a blood sample is taken from a patient and the glucose concentration is directly measured from the blood sample. The finger-prick test can require patients to take a drop of blood from their finger as frequently as 10 times in a day. This can be very uncomfortable and inconvenient for patients over time.

Monitoring devices, such as flash glucose monitors and continuous glucose monitors, may be used when patients require long-term blood glucose monitoring. Such devices are worn just under the skin and measure the amount of glucose in the interstitial fluid, which surrounds body cells. However, such devices must be injected into the body and require patients to wear such devices all the time. Other devices include skin patches including micro-needles that permanently penetrate the skin for continuous monitoring. Such under-skin devices can be uncomfortable, risk infection as well as blood contamination, and have a limited life-span. This is because the human body starts building a protein wall around the sensor after insertion, which disables the sensor and requires the sensor to be replaced after about two weeks.

In view of these currently available means for taking blood glucose measurements, there is a need to provide a non-invasive means of taking glucose measurements in a manner that is comfortable for the user, and resource-efficient. It is an object of the present invention to overcome at least one of the problems in the prior art.

SUMMARY

According to a first aspect of the invention, there is provided a method for determining a dielectric property of a sample, whereby the method is of a system comprising an oscillator circuit including a device for physically coupling with a sample. The method comprises: generating, by the oscillator circuit, at least one electromagnetic signal; bringing a sample into proximity or contact with the device, so as to physically couple the sample with the device; measuring a frequency of the at least one electromagnetic signal output by the oscillator circuit; and calculating a dielectric property of the sample based on the measured frequency of the electromagnetic signal output by the oscillator circuit.

The method may further comprise: measuring a frequency of the electromagnetic signal output by the oscillator circuit when the device is not physically coupled with the sample; determining a shift in the measured frequency of the electromagnetic signal output by the oscillator circuit in response to the sample physically coupling with the device; and calculating a dielectric property of the sample based on the shift of the measured frequency of the electromagnetic signal output by the oscillator circuit.

The method may further comprise: generating, by the oscillator circuit, at least a first electromagnetic signal of a first frequency and a second electromagnetic signal of a second frequency that is higher than the first frequency; determining, in response to the sample physically coupling with the device, a shift in the measured frequency of the first electromagnetic signal output by the oscillator circuit and a shift in the measured frequency of the second electromagnetic signal output by the oscillator circuit; calculating, based on the shift in the measured frequency of the first electromagnetic signal, a level of physical coupling between the sample and the device; obtaining a calibration factor, based on the level of physical coupling between the sample and the device; and calculating a dielectric property of the sample, based on the calibration factor and a measured frequency of the second electromagnetic signal output by the oscillator circuit while the device is physically coupled to the sample.

The dielectric property may comprise permittivity.

At least one of the first and second frequencies may be in the radio frequency band. However, the first and second frequencies may be in other electromagnetic frequency bands, such as the microwave frequency band.

The bringing the sample into proximity or contact with the device may comprise applying, using the sample, pressure to the device. The calculating the level of physical coupling between the sample and the device may comprise calculating the pressure applied by the sample to the device.

The sample may comprise at least one of animal material and plant material. The sample may include a digit of an animal, a digit of a human such as a human finger. The sample may include blood, bone, skin, and fats. The sample may include foods such as fruits and vegetables, as well as drinks.

The method may further comprise determining, based on the calculated dielectric property of the sample, a glucose concentration of the sample. However, the concentration of various constituents other than glucose may be derived from the dielectric property of the sample. For example, when the sample is a human finger, the concentration of various blood constituents may be calculated, such as the concentration of other types of sugars, proteins, red blood cells, and white blood cells. Other biomedical measurements such as blood pressure and respiratory rates may also be derived from the permittivity measurements. Additionally, when the sample is fruits and/or vegetables, the method may include measuring food ripeness and various food contents, such as the levels of fats in milk. Furthermore, the method may be used to determine how safe food is to eat, and/or how spoiled it is.

The first frequency may be lower than 100 MHz. The first frequency may be lower than 10 MHz. The second frequency may be higher than 500 MHz. The second frequency may be higher than 1 GHz. When the first frequency is lower than 100 MHz, the second frequency may be higher than 500 MHz. When the first frequency is lower than 100 MHz, the second frequency may be higher than 1 GHz. When the first frequency is lower than 10 MHz, the second frequency may be higher than 500 MHz. When the first frequency is lower than 10 MHz, the second frequency may be higher than 1 GHz.

At least one of the shifts in the measured frequency of the first electromagnetic signal and the shift in the measured frequency of the second electromagnetic signal may comprise a change in at least one of an amplitude, frequency, phase and delay relationship of the at least one of the first electromagnetic signal and the second electromagnetic signal.

The method may comprise generating, by the oscillator circuit, electromagnetic signals of more than two frequencies.

The device may be a capacitive device. Physically coupling the sample with the device may change the capacitance of the device.

According to a second aspect of the invention, there is provided a device for physically coupling with a sample for use in the method described above.

According to a third aspect of the invention, a system is provided for determining a dielectric property of a sample. The system comprises: an oscillator circuit configured to generate at least one electromagnetic signal; a device for physically coupling with the sample integrated in the oscillator circuit; a measurement device configured to measure a frequency of the at least one electromagnetic signal output by the oscillator circuit; and a processor configured to: determine, in response to the sample physically coupling with the device, a measured frequency of the electromagnetic signal output by the oscillator circuit; and calculate a dielectric property of the sample based on the measured frequency of the electromagnetic signal output by the oscillator circuit.

It may be that the measurement device is configured to measure a frequency of the electromagnetic signal output by the oscillator circuit when the device is not physically coupled with the sample; and the processor is configured to: determine a shift in the measured frequency of the electromagnetic signal output by the oscillator circuit in response to the sample physically coupling with the device, and calculate a dielectric property of the sample based on the shift of the measured frequency of the electromagnetic signal output by the oscillator circuit.

The oscillator circuit may be configured to generate at least a first electromagnetic signal of a first frequency and a second electromagnetic signal of a second frequency that is higher than the first frequency.

The processor may be configured to: determine, in response to the sample physically coupling with the device, a shift in the measured frequency of the first electromagnetic signal output by the oscillator circuit and a shift in the measured frequency of the second electromagnetic signal output by the oscillator circuit; calculate, based on the shift in the measured frequency of the first electromagnetic signal, a level of physical coupling between the sample and the device; obtain a calibration factor, based on the level of physical coupling between the sample and the device; and calculate a dielectric property of the sample, based on the calibration factor and a measured frequency of the second electromagnetic signal output by the oscillator circuit while the device is physically coupled to the sample.

The dielectric property may comprise permittivity.

At least one of the first and second frequencies may be in the radio frequency band.

The processor may calculate the level of physical coupling between the sample and the device based on pressure applied by the sample to the device.

The sample may comprise at least one of animal material and plant material. The sample may comprise a digit of an animal, and the sample may comprise a digit of a human.

The system may comprise a receiving means for receiving a sample, wherein the receiving means comprises a housing for the device, and the housing comprises a base and a lid, and a connector connecting the lid to the base.

The processor may be further configured to determine, based on the calculated dielectric property of the sample, a glucose concentration of the sample.

At least one of the shift in the measured frequency of the first electromagnetic signal and the shift in the measured frequency of the second electromagnetic signal may comprise a change in at least one of an amplitude, frequency, phase and delay relationship of the at least one of the first electromagnetic signal and the second electromagnetic signal.

The oscillator circuit may comprise a negative resistance oscillator. The oscillator circuit may comprise a phase-locked-loop, PLL, circuit.

The measurement device may comprise an Analogue to Digital Converter, ADC, for sampling oscillation frequencies.

The processor may comprise a spectral analyser configured to determine the shift in the measured frequency of the first electromagnetic signal output by the oscillator circuit and the shift in the measured frequency of the second electromagnetic signal output by the oscillator circuit.

The processor may comprise processing circuitry for calculating at least one of the level of physical coupling between the sample and the device, the calibration factor, and the dielectric property of the sample.

The processor may comprise a computing device including the processing circuitry. The computing device may comprise a microcontroller.

The device may have at least one dominant resonance frequency.

The oscillator circuit may be arranged to generate electromagnetic signals of more than two frequencies.

The present invention provides a non-invasive means for determining the concentration of a constituent in a sample of animal and/or plant material. In practice, a dielectric property can be determined by bringing a sample, such as a patient's finger, into contact with the device of the invention. By determining a dielectric property of the sample, various properties of a sample, such as the glucose concentration of a blood sample, can then be derived from the dielectric properties of the sample. In doing so, the present invention is convenient, and user-friendly.

Furthermore, the present invention provides an accurate, efficient and comfortable means for taking blood glucose measurements that minimises human error by virtue of the calibration means. This is particularly beneficial, since, in practice, there is no need to precisely control the amount of pressure that is applied to the device. In doing so, the present invention is also accurate and cost-effective. Moreover, the invention is wide-reaching in application across various industries, including biomedical, food and agricultural sectors.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a schematic view of circuitry in a device according to a first embodiment of the invention; Fig. 2 is a flow chart of a first method according to the invention; and Fig. 3 is a schematic graph of the results of the first method of the invention.

DETAILED DESCRIPTION

Radio frequency (RF) sensors can be used to determine the permittivity of various samples by measuring the dielectric constant of those samples. RF sensors can be designed to be resonant so as to exhibit a strong resonance at a specific frequency. Normally, air, which has a dielectric constant approximately equal to 1, is present on an RF sensor, such that the sensor tends to resonate at its natural resonant frequency. When a sample with a dielectric constant greater than 1 comes into close proximity or contact with the sensor, the sensor's resonance frequency shifts in the frequency spectrum due to the electromagnetic fields radiated from the sensor interacting with the sample. In particular, when a sample is placed on top of the RF sensor, factors such as the permittivity of the sample will alter the electric field radiated by the sensor, which in turn changes the sensor's characteristics, manifests as a change in the sensor's capacitance.

In a blood sample, the permittivity of blood is affected by glucose concentration: as the glucose concentration increases, the permittivity decreases. This is because the ability for blood to store an electric field depends on its glucose concentration. For example, for pure glucose dissolved in de-ionised water, the permittivity difference between a glucose concentration of 0 mg/dl and 500 mg/dl is about 0.51 at 8 GHz and about 0.23 at 1 GHz. The difference in permittivity is less than 0.05 at 100 MHz and becomes even smaller at lower frequencies. In particular, the inventors have identified that the change in permittivity as measured at low frequencies becomes negligible. This is particularly important to take into account for blood glucose measurement, since the expected frequency shift in the normal blood glucose range (4.0 mmol/L to 5.9 mmol/L) is less than 10 MHz at a frequency of about 1 GHz, depending on the sensitivity of the sensor. Accordingly, at lower frequencies, the expected frequency shift becomes so small that a meaningful measurement of the frequency shift cannot be taken.

The inventors have further identified that the quality of the physical coupling between the sample and the sensor also impacts the capacitance of the sensor, and therefore causes a frequency shift in the sensor's frequency response. In particular, the inventors have identified that in the presence of a non-uniform sample such as a human fingertip, this effect on the frequency shift is greater than for uniform samples having smooth and flattened surfaces. This is because human fingertips feature fingerprints consisting of ridges and valleys with inherent air gaps in the valleys. Moreover, fingertips and fingerprints vary in shape and size from person to person, with air gaps varying accordingly from user to user. As the pressure applied onto the sensor by the fingertip is varied, the size and volume of these air gaps change, which alters the effective dielectric constant read by the sensor. Accordingly, changes in pressure introduce a frequency shift in the sensor's frequency response. For example, the frequency shift resulting from applying different levels of pressure can be as high as 200 MHz. However, the frequency shift resulting from a change in pressure may be monitored at any frequency.

The inventors have identified that there is therefore a need to precisely control the pressure in order to obtain meaningful and repeatable results from the measurements by the RF sensor. In the absence of pressure control, the frequency shift resulting from glucose concentration change is masked by the frequency shift resulting from the different level of pressure applied onto the sensor in each measurement, since the frequency shift resulting from glucose measurement is several orders of magnitude smaller than the frequency shift resulting from the pressure applied to the sensor.

In light of the above, the inventors have developed the present invention.

First embodiment A first embodiment of the invention is a device for receiving a sample for non-invasively and accurately measuring blood glucose levels. The sample may be a body part. The body part may be a digit, such as a finger, or an ear lobe, or any other suitable body part. The device comprises a housing for receiving a sample of a user, the housing comprising a base and a lid, and a connector connecting the lid to the base. However, the device may be in any form capable of receiving a sample, and may be provided as a clip having a pair of jaws, such as a pair of elongated portions, which are hinged together by a hinge at one end, so as to clamp down on the received sample. Alternatively, the device may comprise a substantially planar or flat portion having a receiving surface upon which a sample can be placed. For example, when the sample is a finger, the area of the receiving surface is approximately 10 cm', but is not limited to that size.

The device includes circuitry 10, the first embodiment of which is shown in Fig. 1. The circuitry 10 forms part of a circuit board disposed in the device, and spans the area for receiving the sample. The circuitry 10 includes an RF sensor 12 integrated within a multi-frequency RF oscillator, which is connected to a frequency measurement device 14 disposed externally of the device. However, the invention is not restricted to such a configuration. For example, in other embodiments, the frequency measurement device 14 is integrated within the circuit board of the device 10.

The oscillator is configured to generate a plurality of oscillations having predetermined oscillation frequencies, including at least two different oscillation frequencies, including a low frequency and a high frequency. The oscillation having a low frequency of the first embodiment is in the radio frequency band, and is kept at under 100 MHz, preferably between 10 kHz to 20 MHz and more preferably at about 10 MHz. The oscillation having a high frequency is in the radio or microwave frequency band, and is kept over 500 MHz, preferably at least at 1 GHz. There may be more than two frequencies generated. There may be a plurality of high oscillation frequencies and a plurality of low oscillation frequencies that are generated. The RF sensor 12 is a passive device that resonates at a predetermined frequency. The RF sensor 12 may comprise a transducer.

The oscillator in the first embodiment comprises a negative resistance oscillator, which comprises an active device, a negative resistance network providing negative feedback, and an input network. The input network resonates at multiple frequencies. The input network may be integrated with the RF sensor, but in other embodiments, may be provided as a separate network connected to the RF sensor. The RF sensor is designed to be resonant so that it inherently resonates at one or more frequencies. The active device is connected to the input network via the negative resistance network, so as to create a feedback loop that starts the oscillations. These oscillations are then amplified by the active device and the resulting amplified signal is output to the frequency measurement device via the output network.

The active device in the first embodiment comprises a transistor, such as a bipolar junction transistor (BJT) or MOSFET. The active device is used to generate the plurality of oscillations, by operating in its unstable region of operation. The load or source impedance can be predetermined in the unstable impedance regions to exhibit negative resistance, such that when the voltage increases, the current decreases. In doing so, the oscillator performs as a destabilised amplifier. The active device 16 has a first port 24, a second port 26 and a third port 28. The first port 24 is connected to a DC feed network 30 and the output network 22. The second port 26 is connected to the input network 20, and the third port 28 is connected to the negative resistance network 18. When the active device 16 is a BJT, the first, second and third ports 24, 26, 28 are a collector 24, a base 26 and an emitter 28, respectively. When the active device 16 is a MOSFET, the first, second and third ports 24, 26, 28 are a drain 24, gate 26 and source 28, respectively.

The input network 20 and the output network 22 comprise predetermined selected networks of inductors and capacitors that facilitate impedance matching between the constituent components of the circuitry 10. More specifically, the input network 20 provides impedance matching between the RF sensor 12 and the port 26. The output network 22 provides impedance matching between the frequency measurement device 14 and port 24. By matching impedance in this way, the circuitry 10 ensures maximum power transfer between the RF sensor 12 and the frequency measurement device 14 so that no power or negligible power is reflected back. In doing so, signal distortion may be minimised, thereby providing enhanced measurement accuracy and reliability. This is because the negative resistance network 18 provides a feedback loop between the second port 26 and the third port 28. In response to this feedback loop together with the predetermined capacitance and inductance of the RF sensor 12, the input network 20 by the second port 26 starts oscillating. These oscillations are observed at the first port 24, and output to the frequency measurement device 14.

The RF sensor 12 is a passive device that can have one or more resonant frequencies, and may have a printed circuit board (PCB) laminate. The RF sensor 12 and the oscillator are optimised for integration with one another so as to define and optimise the resonance frequency of the sensor 12. For example, this optimisation can be achieved using the oscillator components by predetermining the oscillator topology and parameters, such as the materials and thickness of the PCB laminate, as well as other characteristics. This optimisation can also be achieved by designing the RF sensor 12 to increase the capacitance of the RF sensor, for example by predetermining the capacitance and impedance of the RF sensor 12. However, other suitable means may be used to optimise the RF sensor 12 and oscillator for integration with one another.

A first method using the device described above for measuring blood glucose levels will now be described. Fig. 2 shows the steps of the first method in a flow chart.

In step S100, the device is set up in the manner described above so as to generate oscillations. More specifically, the oscillator generates oscillations having predetermined radio and microwave frequency band frequencies, whereby at least one of the oscillation frequencies is a low frequency-component fi of for example and not restricted to 10 MHz and at least one oscillation frequency is a high frequency-component f2 of for example and not restricted to 1 GHz. However, the invention is not limited to these frequency components, as the low frequency-component fi may be a radio frequency selected from the range of 10 kHz -100 MHz and the high frequency-component f2 is selected from the radio and microwave range. For example, the high frequency component may be higher than 500 MHz. For example, in some embodiments, the high frequency component f2 is approximately 40 GHz, and in other embodiments, the high frequency component f2 is 60 GHz. However, the invention is not limited to the high frequency components of these embodiments. The frequency measurement device 14 reads the oscillations, including but not limited to those having frequencies f1 and f2, output by the oscillator 12 in step S120.

Following the frequency oscillation readings, a sample is applied to be received by the device, applying pressure on the receiving surface of the device in step S130. The area of the receiving surface of the device upon which the sample is applied may include the RF sensor 12, such that the sample is applied directly to the RF sensor 12. However, the invention is not restricted to this configuration. In other embodiments, the sample is not applied directly onto the RF sensor 12, but may instead be proximal to the RF sensor 12. For example, the sample could hover above the sensor 12 and still affect the permittivity of the sensor, thereby facilitating the invention. It is understood by the skilled person however that the hovering separation between the sample and the sensor is dependent on the sensitivity of the sensor.

The frequency measurements are retaken following the application of the sample to the device in step S140. In response to the sample applying pressure on the device, the low and high frequencies fi and f2 measured at the output network 22 are shifted to lower respective frequencies, fis and f2s. Fig. 3 shows a schematic graph illustrating the shifts in the low and high frequency components resulting from a sample being applied to a device. A shift in the low frequency component fi_is amounts to the difference between the reading of the low frequency fi in the absence of the sample being applied to the device and the reading of the low frequency fls while the sample is being applied to the device. A shift in the high frequency component f2-2s amounts to the difference between the reading of the high frequency f2 in the absence of the sample being applied to the device and the reading of the high frequency f2s while the sample is being applied to the device.

The inventors have identified that the shifts in the frequency components, fi_is and f2.215, arise, as follows. When a sample is brought into contact or proximity with the device described above, the capacitance in the sensor 12 changes in response to a) the quality of the physical coupling between the sample and the sensor 12, and b) the blood glucose concentration in the sample, thereby causing the shifts in frequency spectra, fi_is and f2.2s, shown in Fig. 3. More specifically, the inventors have identified that the shifts in the low and high frequency components, fiiis and f2.2s, are both functions of the physical coupling between the sample and the RF sensor 12. However, as discussed above, only the shift in the high-frequency component, f2-25, is substantially affected by the permittivity of the sample (and therefore its blood glucose concentration). The shift in the low frequency component, is, results primarily from the pressure applied by the sample on the sensor 12, since the impact of the permittivity of the sample on the RF sensor 12 is so small as to be negligible at low frequencies. In embodiments where the sample is in proximity to the sensor, the shift in the low frequency component, f1.1s, results primarily from the physical coupling between the sample and the sensor. The permittivity value can therefore be determined at least approximately by observing and processing the shift in the high frequency component f.2-2S. based on the shift in the low frequency component f1.1s, as follows: The shifts in the low and high frequency components, fiiis and f2.2s, are calculated in step S150. In the first embodiment, the low frequency shift fi_is is used to determine the level of physical coupling/pressure applied to the sensor, which is used in turn to determine the calibration factor. The calibration factor can be arrived at through a mathematical model or a simulation based approach. The calibration factor is applied to the shifted high frequency component, f2s, to obtain a corrected high frequency component, f2c. The difference between the reading of the high frequency component, f2, in the absence of the sample being applied to the device and the corrected shifted high frequency component, f2c, is subsequently calculated, to obtain a corrected shift in the high frequency component, f2_20. In doing so, the effect of the physical coupling between the sample and the sensor 12 is removed from the shift in the high frequency component so as to isolate the effect of glucose on the capacitance of the sensor 12 in the high-frequency component f2. In other words, the corrected shift in the high frequency component, f. 2-2C represents the effect the permittivity of the sample has on the capacitance of the RF sensor 12, without the pressure applied by the sample on the sensor 12 skewing the readings. This is because the shift in the low frequency component, accounts for the physical coupling, whereas the shift in the high frequency component, f2-2s, accounts for both the physical coupling and the permittivity of the sample.

In the first embodiment, the calculations in steps S150 and S160 are performed by a frequency measurement device 14. The frequency measurement device 14 of the first embodiment comprises a processor and a spectral analyser. The spectral analyser may be a Vector Network Analyser (VNA) that has been configured to be a spectral analyser. The spectral analyser analyses the electromagnetic signals output by the oscillator and communicates with the processor. The processor identifies the shifted low and high frequency components, fis and f2s, based on the measured frequency components measured by the spectral analyser, and calculates the shifts, fi.is and f2-25, the calibration factor, the corrected high frequency component, f2c, and the corrected shift in the high frequency component, f. 2-20. The communication between the processor and the spectral analyser can be via any suitable means, such as wired or wireless communication. The processor of the first embodiment further comprises processing circuitry for processing the signals to calculate the calibration factor, the corrected high frequency component, f2c, and the corrected shift in the high frequency component, f2.2c. The processing circuitry may be integrated within a computing device, such as a microcontroller.

In the first embodiment, the processing circuitry of the processor is configured to also derive the dielectric constant of the sample that has been applied to the device, based on the corrected shift in the high frequency component, f. 2-2C in step S170, and subsequently calculates the blood glucose concentration of the sample, based on the derived dielectric constant. In particular, the processor performs processing to translate the measured frequencies into permittivity and blood glucose readings, based on mathematical modelling.

The first embodiment therefore provides a non-invasive, efficient and cost-effective means for determining the dielectric constant of a sample, such as a patient's finger, from which the blood glucose concentration of the sample can subsequently be derived. In particular, the low and high frequency components, fi and f2, are predetermined in order to accurately determine the blood glucose concentration, as follows.

The amount of glucose in a healthy person's blood roughly ranges from 75mg/dI to 110mg/d1. As discussed above, permittivity is a function of frequency, whereby a higher measurement resolution can be obtained at higher frequencies. For example, a higher measurement resolution (and therefore accuracy) can be achieved at 10 GHz instead of 2 GHz. However, as the frequency increases, the penetration depth of the waves decreases, as waves travel less deeply into the skin and the tissues, thereby decreasing the measurement depth and hence the measurement accuracy. Moreover, the size of the sensor 12 can be reduced as the frequency is increased, which helps to maintain the position of the sample on the sensor 12 more easily so as to improve total coverage of the sensor 12 by the sample. Hence, by predetermining the low frequency component, f,, to for example be 10 MHz, the first embodiment ensures that the shift in the low frequency component, f1.18, is not a function of the permittivity of the sample, and further ensures that the size of the sensor 12 is small enough to receive a sample for reliable and repeatable measurements. Moreover, by predetermining the high frequency component, f2, to for example be at least 1 GHz, this ensures that the shift in the high frequency component, f2-25, is high enough to take an accurate permittivity measurement. This is because at frequencies lower than for example about 500 MHz, the shift resulting from blood glucose may not be substantial enough to take an accurate measurement. Hence, the low and high frequency components of the first embodiment ensure that the size of the sensor 12 is practical for providing a suitable and appropriate surface to receive a sample, and the penetration depth is substantial enough to provide a meaningful, reliable and repeatable reading, with a sufficient permittivity resolution.

Moreover, by designing the sensor 12 to be highly capacitive as in the first embodiment, the sensitivity of the sensor 12 to permittivity at the higher frequencies can be optimised, so as to obtain a high permittivity resolution.

Other embodiments However, the invention is not limited to the above described first embodiment, as follows.

In the first embodiment described above, the shifts in the low and high frequency components, fi.is and f2.2s, are in relation to the frequency difference between fi and fis, and f2 and f2s, respectively. However, the invention is not limited to this configuration, since there are several properties in the spectra that can be measured to quantify the physical coupling and the permittivity of the sample. For example, in other embodiments, at least one of the amplitude difference and phase/delay relationship between fl and fis, and f2and f2s may be calculated, instead of, or in addition to, the difference in frequencies between the frequency components.

Additionally, the above described processor of the frequency measurement device 14 is not limited to performing all the above processing of steps S150 to S180, but other different components may be responsible for one or more of these steps. For example, the frequency measurement device may output the calculations to a separate processor that may in turn be configured to determine the dielectric constant and blood glucose concentration of the bsample.

Moreover, the measurement and analysis of the frequencies may be performed using other signal processing techniques in the place of spectral analyser equipment, which may be sought-after in view of the expense that spectral analysers present. For example, in other embodiments, the spectral analyser may be replaced with a phase-locked-loop (PLL) circuit, which locks the frequency components to a known external oscillator frequency and measures the loop error voltage. Hence, the PLL represents a more cost-effective means of measuring the signals, as compared with the spectral analyser of the first embodiment.

In further embodiments, another type of signal processing technique is provided, as follows. The frequency measurement device 14 of the first embodiment may comprise an Analog to Digital Converter (ADC), which is configured to sample the measured signals in order to process them to obtain the frequencies. For example, the shifted low frequency, fis, may be sampled via the ADC, and the shifted high frequency, f2s, may be likewise sampled by the ADC after being down-converted. In doing so, the shifts in the low and high frequency components, f2-2s and f2-2s, can be directly compared and processed, as required. Alternatively, the shifted low frequency, fls, may be sampled via an ADC, and the shifted high frequency, f2s, may be subsequently sampled based on the shifted low frequency, fis, using at least one of direct-sampling, sub-sampling, and over-sampling techniques. These sampling techniques allow a generalised ADC to be used, which does not have to be tailored for use with the operating frequencies of the oscillator and sensor 12 of the device. For example, if the frequencies of the oscillations are not within the range of the ADC's sampling rate, the use of a sub-sampling technique facilitates measurement of the signals using an ADC that would not otherwise be fast enough to sample the signals. Over-sampling can also be utilised to increase the measurement accuracy. In using any of these sampling techniques, it can be beneficial to use more than one low and high frequency component, so as to obtain more measurement results from the sample at different frequencies, thereby leading to an increased accuracy in calculating the permittivity.

Additionally, the oscillator is not restricted to the negative resistance oscillator described in the first embodiment above, and may generate oscillations in other suitable ways. For example, in some embodiments, the oscillator of the first embodiment may be replaced with a Vector Network Analyser (VNA). The VNA is a validation tool that is configured to measure the sensor's frequency response over a predefined range of frequencies to identify the one or more resonance frequencies of the RF sensor. In such embodiments, the VNA sweeps the frequency in the predefined identified range of the one or more resonance frequencies, and subsequently feeds the generated RF signal to the RF sensor and measures the reflected RF power to perform the measurement in that part of the spectrum.

Furthermore, the invention is not limited to the radio frequency bands, but may be implemented using other electromagnetic frequency bands, such as the optical frequency band. It is understood by the skilled person that the oscillations by the active device are suitably configured for use with sensors operating in different electromagnetic frequency bands.

The present invention is also wide-reaching in its application beyond blood glucose measurement. This is because the RF sensors can measure the dielectric constant of many types of samples, and can be designed to target measurements as required. In particular, the frequency oscillations of the sensor respond to any change in dielectric constant in any suitable media. Accordingly, by modifying the oscillator frequencies and the RF sensor design, the invention can in practice be applied to any medium where a property can be derived from a permittivity measurement. For example, across the biomedical industry, the method can be implemented to measure the concentration of various blood constituents other than glucose, such as proteins, red blood cells, and white blood cells, with the appropriate calibration. Furthermore, other types of samples other than blood may be tested upon, such as bone, skin, and fats. Other biomedical measurements such as blood pressure and respiratory rates may also be derived from the permittivity measurements.

The method may also be implemented in other industries, from food to agricultural sectors. For example, when applied to the food industry, the method may be used to measure the glucose concentration of fruit, from which the ripeness of the fruit under inspection can be determined. The method may also be used to measure the ripeness of other foods, as well as food contents, such as the level of alcohol in drinks, and levels of fats in milk.

Furthermore, the method may be used to determined how safe food is to eat, and/or how spoiled it is. When applied to the agricultural sector, the method may be implemented to measure soil humidity, and soil contents such as fertiliser levels. This is particularly beneficial when applied to farming, since the times for feeding plants can be optimised by taking into account the various constituents of soil that are present at a given time.

The foregoing description has been given by way of example only and it will be appreciated by a person skilled in the art that modifications can be made without departing from the scope of the present invention as defined by the claims.

Claims

CLAIMS1. A method for determining a dielectric property of a sample, the method using a system comprising an oscillator circuit including a device for physically coupling with a sample, and the method comprising: generating, by the oscillator circuit, at least one electromagnetic signal; bringing a sample into proximity or contact with the device, so as to physically couple the sample with the device; measuring a frequency of the at least one electromagnetic signal output by the oscillator circuit; and calculating a dielectric property of the sample based on the measured frequency of the electromagnetic signal output by the oscillator circuit.
2. The method according to claim 1, the method further comprising: measuring a frequency of the electromagnetic signal output by the oscillator circuit when the device is not physically coupled with the sample; determining a shift in the measured frequency of the electromagnetic signal output by the oscillator circuit in response to the sample physically coupling with the device; and calculating a dielectric property of the sample based on the shift of the measured frequency of the electromagnetic signal output by the oscillator circuit.
3. A method according to claim 1, the method further comprising: generating, by the oscillator circuit, at least a first electromagnetic signal of a first frequency and a second electromagnetic signal of a second frequency that is higher than the first frequency; determining, in response to the sample physically coupling with the device, a shift in the measured frequency of the first electromagnetic signal output by the oscillator circuit and a shift in the measured frequency of the second electromagnetic signal output by the oscillator circuit; calculating, based on the shift in the measured frequency of the first electromagnetic signal, a level of physical coupling between the sample and the device; obtaining a calibration factor, based on the level of physical coupling between the sample and the device; and calculating a dielectric property of the sample, based on the calibration factor and a measured frequency of the second electromagnetic signal output by the oscillator circuit while the device is physically coupled to the sample.
4. The method of 3 wherein the calculating the level of physical coupling between the sample and the device comprises calculating the pressure applied by the sample to the device.
5. The method of claim 3 or claim 4, wherein the first frequency is lower than 100 MHz, and preferably lower than 10 MHz, and wherein the second frequency is higher than 500 MHz, and preferably higher than 1 GHz.
6. The method of any one of claims 3 to 5, wherein at least one of the first and second frequencies is in the radio frequency band.
7. The method of any one of the preceding claims, wherein the bringing the sample into proximity or contact with the device comprises applying, using the sample, pressure to the device.
8. The method of any one of the preceding claims, wherein the sample comprises at least one of animal material and plant material.
9. The method of claim 8, wherein, when the sample comprises animal material, the sample comprises a digit of an animal, and optionally the sample comprises a digit of a human.
10. The method of any one of the preceding claims, further comprising determining, based on the calculated dielectric property of the sample, a glucose concentration of the sample.
11. The method of any one of the preceding claims, wherein at least one of the shift in the measured frequency of the first electromagnetic signal and the shift in the measured frequency of the second electromagnetic signal comprises a change in at least one of an amplitude, frequency, phase and delay relationship of the at least one of the electromagnetic signals.
12. A system for determining a dielectric property of a sample, the system comprising: an oscillator circuit configured to generate at least one electromagnetic signal; a device for physically coupling with the sample integrated in the oscillator circuit; a measurement device configured to measure a frequency of the at least one electromagnetic signal output by the oscillator circuit; and a processor configured to: determine, in response to the sample physically coupling with the device, a measured frequency of the electromagnetic signal output by the oscillator circuit; and calculate a dielectric property of the sample based on the measured frequency of the electromagnetic signal output by the oscillator circuit.
13. The system of claim 12, wherein: the measurement device is configured to measure a frequency of the electromagnetic signal output by the oscillator circuit when the device is not physically coupled with the sample; and the processor is configured to: determine a shift in the measured frequency of the electromagnetic signal output by the oscillator circuit in response to the sample physically coupling with the device, and calculate a dielectric property of the sample based on the shift of the measured frequency of the electromagnetic signal output by the oscillator circuit.
14. The system of claim 12, wherein: the oscillator circuit is configured to generate at least a first electromagnetic signal of a first frequency and a second electromagnetic signal of a second frequency that is higher than the first frequency; and the processor is configured to: determine, in response to the sample physically coupling with the device, a shift in the measured frequency of the first electromagnetic signal output by the oscillator circuit and a shift in the measured frequency of the second electromagnetic signal output by the oscillator circuit; calculate, based on the shift in the measured frequency of the first electromagnetic signal, a level of physical coupling between the sample and the device; obtain a calibration factor, based on the level of physical coupling between the sample and the device; and calculate a dielectric property of the sample, based on the calibration factor and a measured frequency of the second electromagnetic signal output by the oscillator circuit while the device is physically coupled to the sample.
15. The system of claim 14, wherein the processor is configured to calculate the level of physical coupling between the sample and the device based on pressure applied by the sample to the device.
16. The system of claim 14, wherein the first frequency is lower than 100 MHz, and preferably lower than 10 MHz, and wherein the second frequency is higher than 500 MHz, and preferably higher than 1 GHz.
17. The system of any one of claims 12 to 16, wherein at least one of the first and second frequencies is in the radio frequency band.
18. The system of any one of claims 12 to 17, wherein the sample comprises at least one of animal material and plant material.
19. The system of claim 18, wherein, when the sample comprises animal material, the sample comprises a digit of an animal, and optionally the sample comprises a digit of a human.
20. The system of claim 19, further comprising a receiving means for receiving a sample, wherein the receiving means comprises a housing for the device, and wherein the housing comprises a base and a lid, and a connector connecting the lid to the base.
21. The system of any one of claims 12 to 20, wherein the processor is further configured to determine, based on the calculated dielectric property of the sample, a glucose concentration of the sample.
22. The system of any one of claims 12 to 21, wherein at least one of the shift in the measured frequency of the first electromagnetic signal and the shift in the measured frequency of the second electromagnetic signal comprises a change in at least one of an amplitude, frequency, phase and delay relationship of the at least one of the first electromagnetic signal and the second electromagnetic signal.
23. The system of any one of claims 12 to 22, wherein the oscillator circuit comprises a negative resistance oscillator or a phase-locked-loop, PLL, circuit.
24. The system of any one of claims 12 to 23, wherein the measurement device comprises an Analogue to Digital Converter, ADC, for sampling oscillation frequencies.
25. The system of any one of claims 12 to 24, wherein the processor comprises a spectral analyser configured to determine the shift in the measured frequency of the first electromagnetic signal output by the oscillator circuit and the shift in the measured frequency of the second electromagnetic signal output by the oscillator circuit.