Method of determining glucose concentration in tissue, computer program product, radiometer and a system therefor

Abstract

A method of non-invasive determination of the blood glucose concentration in the patient's tissue based on a radio noise signal received using an antenna brought close to the patient's skin according to the invention involves use of the radio noise signal is measured using a passive radiometer and additionally the method includes the following steps: a step of obtaining transformation coefficients, a step of measuring the temperature of the tissue surface, a step of measuring the temperatures of the active elements of the receiving chain of the radiometer, a step of measuring the currents consumed by the active elements of the receiving chain of the radiometer, a step of measuring the power of the radio noise signal originating from the tissue, and a step of determining the blood glucose concentration based on the aforementioned values. The invention also relates to a computer program, a radiometer, and a device for determining glucose concentration.

Description

Method of determining glucose concentration in tissue, computer program product, radiometer and a system thereforField of the invention

[0001] The invention concerns a method for determining the glucose concentration in tissue, a computer program product, a radiometer, and a system for determining the glucose concentration in tissue, especially in the blood.State of the art

[0002] Determining the blood glucose concentration, i.e., the so-called glycaemia, is important due to the need for monitoring glucose levels related to the treatment and prevention of diabetes .

[0003] Diagnosing diabetes based on a blood test gives a precise result. However, for the patient it is associated with a negative stimulus, i.e., pain associated with the invasive examination, so work on alternative solutions is constantly ongoing. This applies especially to mobile devices. Various techniques are used, including iontophoresis-based, ultrasound, detection of respiratory biomarkers, temperature measurement, examination of tissues with electromagnetic waves and light, examination of radiation reflected from tissues, and others.

[0004] In the field of measurements with the use of microwave radiometer technology, there are many solutions of radiometer systems for measuring the temperature inside tissues for the diagnosis of cancer, measuring core temperature, diagnosing vesicoureteral reflux, measuring brain temperature, etc. Passive radiometers are devoid of the transmitters and are device for measuring the radiant flux (power) of electromagnetic radiationin particular noise. Accordingly they are applicable for indirect measurement of temperature e.g. MDPI and ACS Style Park, W.; Jeong, J. Total Power Radiometer for Medical Sensor Applications Using Matched and Mismatched Noise Sources. Sensors 2017, 17, 2105. https: / / doi.org / 10.3390 / sl7092105. Total power radiometer is a simple construction with a cascade of amplifiersin receiving chain and detector for direct measurement of the power of the incoming (noise ) signal . Usually the power is scaled into a temperature units - so called brightness temperature . Total power radiometers can be used in more complex configurations of the receiving including switches and appliances for reference measurements . Advanced constructions comprising passive total power radiometers involve Dicke radiometer and correlation radiometer K . . These constructions are based on using reference from standardi zed reference noise source . Dicke radiometer sequentially measures the input signal and reference source with one switched receiver chain . Correlation radiometer comprises two identical receiving chains based on total power radiometers . There are also other constructions of passive radiometers based on one or more passive radiometer .

[0005] From the European patent application EP1949084 , a device for minimally invasive measurement of the concentration of components contained in the structure of biological tissue is known, the device comprising a microwave energy source generating a range of microwave frequencies , a first antenna connected to the microwave energy source and adapted to transmit at least a portion of the microwave energy to the tissue structure , a second antenna adapted to receive at least a portion of the microwave energy transmitted through the tissue structure , a signal processor adapted to determine a resonant frequency of the received microwave energy, and a data processor adapted to provide an output of the concentration of components in the biological tissue structure on the basis of the determined resonant frequency . US2016051171 discloses a device for interrogating human skin using tight coupling between the transmitter and receiver of the millimeter waves . It uses methods to evaluate changes in the amplitude and / or phase of the transmitted the millimeter waves in order to estimate the blood concentration of glucose . DE102006055691 discloses a device witha diagnostic measuring unit for generating measuring signals , and an evaluating unit for determining a physiological parameter by processing the measuring signals . The measuring unit is integrated in a keyboard of a computer e . g . laptop . The measuring unit is an optical measuring unit for generating oximetric and / or plethysmographic measuring signals . The evaluating unit is provided for determining a local metabolic parameter e . g . local oxygen usage , from the signals of the optical measuring unit . WO2019147582 . This document discusses , among other things , systems and methods to compensate for the ef fects of temperature on sensors , such as analyte sensor . An example method may include determining a temperature-compensated glucose concentration level by receiving a temperature signal indicative of a temperature parameter of an external component , receiving a glucose signal indicative of an in vi vo glucose concentration level , and determining a compensated glucose concentration level based on the glucose signal , the temperature signal , and a delay parameter .

[0006] CN110074790 discloses a method of non-invasive determination of the blood glucose concentration based on a blood layer in an earlobe using a time-domain processed microwave signal . The method includes the following steps : creating an earlobe model ; preparing test solutions with di f ferent glucose concentrations to simulate blood sugar concentration; placing two antennas on two sides of the earlobe model , where the transmitting antenna emits a high- frequency sinusoidal signal and the receiving antenna receives the sinusoidal s ignal passing through the simulated earlobe ti ssues . Using glucose solutions of di f ferent concentrations during measurement and signal reception allows one to determine calibration coef ficients . The received signals are analyzed in the time domain, wherein the analysis includes filtering the noise from the signals using a wavelet trans form algorithm, analyzing the trend of the filtered received signals with the change in glucose concentration,obtaining the rule of the amplitude of the energy density spectrum of the signals after filtering the noise as a function of the glucose concentration value .

[0007] Solutions based on exposing the patient to radiation ( infrared, light , or radio waves ) or ultrasound have the disadvantage that they are not always indi f ferent to the patient . This especially applies to children under 2 years of age and pregnant women . Occasional exposure does not have a maj or impact on patients , but constant exposure , such as when the blood glucose concentration is measured using a wearable device , may not be indi f ferent and veri fication of the consequences of this exposure requires long-term research .

[0008] From the European patent application EP1224905 , it is known to use a total power radiometer in temperature measurement , wherein the radiometer operates in the 400 MHz or 100 MHz band at center frequencies of 1 . 5 GHz or 3 . 5 GHz and analyses noise radiation generated by tissues to determine temperature .

[0009] In the MSc thesis "A Microwave Radiometer System for Use in Biomedical Applications" , Laura Ballew, Baylor University, 2006 , the use of a microwave radiometer to determine glycemia was proposed . This solution is quite complicated and requires a precise noise source with a stable temperature to obtain a radiometer in the Dicke configuration .

[0010] Various radiometer designs are typically used : Total Power Radiometers ( TPR) , Dicke radiometers , and correlation radiometers .

[0011] The disadvantage of TPRs is the sensitivity of the gain to temperature changes . The advantages include simple structure — it is basically a cascade of ampli fiers with a detector placed at the end . Its advantage is simplicity, while its disadvantage is susceptibility to changes in the gain and changes in noise characteristics resulting from changes in the temperature of its elements . This is problematic because temperature changes depend not only on the surroundings , but also on the amount of heatreleased in the elements of the receiving chain due to bias power consumption .

[0012] The Dicke radiometer requires the use of additional noise sources with precisely stabilized temperature and microwave components in the form of a circulator and a switch to enable gain control. In the Dicke configuration a single total power radiometer is switched between two input signals: measured one and reference noise. Such a design reduces sensivity to changes in the gain and temperature. Example of Dicke radiometer is disclosed in Yan Li, Liang Lang, Qingxia Li, Siyuan Liu and Liangqi Gui, "A novel Dicke microwave radiometer without temperature control for reference match load," 2016 IEEE International Conference on Microwave and Millimeter Wave Technology (ICMMT) , Beijing, China, 2016, pp . 880-882, doi: 10.1109 / ICMMT .2016.7762473. Another example is known from J. Lee, G. S. Botello, R. Streeter, K. Hall and Z. Popovic, "A Hybrid Correlation-Dicke Radiometer for Internal Body Thermometry, " 2022 52nd European Microwave Conference (EuMC) , Milan, Italy, 2022, pp . 464-467, doi: 10.23919 / EuMC54642 .2022.9924276.

[0013] The correlation radiometer requires the use of two total power radiometers and additional elements in the form of microwave couplers, several detectors (from 2 to 8) and a reference noise source. Such a design ensures insensitivity to changes in the gain. Example of correlation radiometer is disclosed in . Tanner, J. Bosch-Lluis and P. Kangaslahti, "Applications of the Pseudo-Correlation Microwave Radiometer, " IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium, Kuala Lumpur, Malaysia, 2022, pp . 7241-7244, doi : 10.1109 / IGARSS46834.2022.9883145.Problem to be solved

[0014] The problem is to find a passive and non-invasive method for determining the glucose concentration in tissue based on measurement involving a simple apparatus of construction thatallows miniaturi zation and is applicable to simple electronic circuitry — not only portable , but also wearable .Summary of the invention

[0015] A computer implemented method of non-invasive determination o f the blood glucose concentration in the patient ' s tissue based on a radio noise signal , especially a microwave noise signal , received using an antenna brought close to the patient ' s skin, according to the invention is distinguished by that the radio noise signal is received using a passive radiometer advantageously a total power passive radiometer . Passive radiometers do not require transmitter or source but are adapted to measure low power noise signals emitted by obj ects . Additionally, the method includes the following steps : a step of obtaining trans formation coef ficients , a step of measuring the temperature of the tissue surface , a step of measuring the temperatures of the active elements of the receiving chain of the radiometer, a step of measuring the currents consumed by the active elements of the receiving chain of the radiometer, a step of measuring the power of the radio noise signal originating from the tissue , and a step of determining the blood glucose concentration in the tissue based on the trans formation coef ficients and the measured values of the temperature of the tissue surface , the temperatures and the currents consumed by the active elements of the receiving chain of the radiometer, and the power of the radio noise signal originating from the tissue . The total power radiometer has an uncomplicated design that is easy to miniaturi ze . Determining the glucose concentration based on the noise power requires determining trans formation coef ficients to establish a relationship between the power of the noise generated by the tissue and the ef fect of the glucose concentration on the antenna mismatch . The glucose concentration af fects impedance of the tissue and therefore af fects impedance matching of the antenna . The coef ficients can be determined in reference measurements . Inreference measurements , it is also possible to determine the trans formation coef ficients for the receiving chain i f the temperatures of its elements and the currents consumed by them are measured . In this way, it is possible to use a simple , miniaturi zed design . Method is applicable also for improving accuracy of measurements with di f ferent kinds of radiometers .

[0016] Advantageously, the method is implemented in a computer, and in the step of obtaining trans formation coef ficients , the predefined trans formation coef ficients are loaded from the computer memory or downloaded from an external server .

[0017] Advantageously, the trans formation coef ficients are determined based on a series of measurements performed on a set of tissues or phantoms with known glucose concentrations , the measurements being performed at di f ferent temperatures .

[0018] Advantageously, the trans formation coef ficients are determined in the process of multivariate regression consisting in minimi zing the error of matching glucose concentration reference values to the values calculated from the transient characteristics of the radiometer, wherein the variables are : the temperature of the tissue surface , the temperatures of the active elements of the receiving chain of the radiometer, the currents consumed by the active elements of the receiving chain of the radiometer, the power of the radio noise signal . The easiest way is to use the mean squared error criterion .

[0019] Advantageously, converting the measured values and the trans formation coef ficients into the blood glucose concentration includes a step of determining the power of the noise received by the antenna based on the trans formation coe f ficients , the power measured using said radiometer, the temperatures of the active elements of the receiving chain, and the currents consumed by the active elements of the receiving chain of the radiometer . Next , the power of the noise originating from the tissue and reaching the antenna and the reflection coef ficient at the interface between the antenna and the tissue is determined usingthe previously determined power of the noise received by the antenna, the trans formation coef ficients , and the measured temperature of the tissue surface . Finally, the glucose concentration is determined using the determined reflection coef ficient at the interface of the antenna and the tissue and the determined power of the noise originating from the tissue and reaching the antenna .

[0020] Advantageously, the trans formation coef ficients are the weight coef ficients of the neural network processing the data set containing the measured temperatures , the currents consumed by the active elements of the receiving chain of the radiometer, and the power of the radio noise signal originating from the tissue .

[0021] Advantageously, the measurements of the radio noise signal are carried out in the band ranging from 800 MHz to 24 GHz . Measurements in this frequency range are reliable and have been proven to be repeatable .

[0022] Advantageously, the measurement is carried out in a band with a width greater than or equal to 200 MHz , and even more advantageously in a band with a width greater than or equal to 500 MHz . This allows the measurement to be performed faster and the influence of patient movements is eliminated .

[0023] A computer program product to be run on a computer adapted to receive measurement data according to the invention comprises instructions that causes the method according to the invention to be performed .

[0024] A radiometer according to the invention comprises an antenna and a radio receiving chain with a detector . The antenna is adapted to be applied to the skin of the patient and to receive a radio noise signal from subcutaneous tissue . The radiometer is a passive radiometer adapted to measure power of noise signal originating from the tissue . The radiometer is provided with sensors of the temperatures of the active elements of the receiving chain, a sensor of the patient temperatureplaced on the antenna, and sensors of the currents consumed by the active elements of the receiving chain of the radiometer .

[0025] Advantageously, the antenna has a frequency band at least 20% wider than the frequency band of the receiving chain . Owing to this , the device is resistant to the dri ft of the propagation parameters of the skin and the related shi ft of the antenna operating band .

[0026] Advantageously, the receiving chain of the radiometer contains a cascade of ampli fiers separated by isolators .

[0027] Advantageously, the receiving chain contains a cascade of ampli fiers separated by attenuators with a rated attenuation below 3 dB .

[0028] A system for determining glucose concentration based on a radio signal , comprising a processing system and a radiometer with an antenna adapted to be applied to the patient ' s skin, according to the invention is distinguished by that the radiometer is a radiometer according to the invention, and the processing system connected thereto : sensors of the temperatures of the active elements of the receiving chain of the radiometer, a sensor of the temperature of the tissue surface placed on the antenna, and current measurement modules measuring the currents consumed by the active elements of the receiving chain of the radiometer, and the processing system is adapted to carry out the method according to the invention . Such a system can be miniaturi zed and can be used as a wearable system while ensuring high accuracy .Description of the drawing

[0029] The subj ect of the invention is described in embodiments in the drawing, in which Fig . 1 shows a block diagram of a system for determining glucose concentration according to the invention .Description of embodiments

[0030] According to the invention, a passive solution is used . No signal is transmitted to the tissues , but a total power radiometer measures the noise signal originating from the tissues .

[0031] The total power radiometer is a radio receiver containing an antenna and a radio receiving chain with a detector shown in the block diagram in Fig . 1 . The antenna 2 is adapted to be applied to the skin 1 of the patient and to receive a radio noise signal from subcutaneous tissue .

[0032] The antenna is a wideband antenna having the operating band of approximately 1 GHz , which is wider than the operating band of the radiometer chain, which in this example equals 300 MHz . Owing to thi s , the entire system works even in the presence of a dri ft of the antenna operating band caused by a change in the electrical parameters of the skin, mainly caused by a change in skin moisture as a result of moisture released due to sweating or TEWL ( Transepidermal Water Loss ) . In this embodiment , antenna 2 is made using not symetric strip - so called microstrip - technology with dimensions and electrical permittivity of the substrate selected to enable the reception of the radiation originating from skin 1 . The microstrip has the advantage that the antenna structure itsel f shields interfering signals not originating from the patient ' s skin . A person skilled in the art is also able to use other antenna designs that will ensure the reception of the signal from the tissue .

[0033] Antenna 2 is connected directly to the radiometer chain, consisting of a cascade of four ampli fiers 3, 5 , 7 , 9 with a gain of approximately 20 dB, a bandpass filter 10 with a bandwidth of 300 MHz , and a diode detector 11 at the end of the cascade . The output of the detector 11 is the voltage at its terminals representing the measured noise signal power . A person skilled in the art is also able to use other detectors that will ensure determining the signal strength .

[0034] Preferably, microwave isolators 4 , 6, 8 are placed between the ampli fiers . I solators with an insertion loss of 0 . 2 dB and an isolation of 20 dB are used . Such a design makes it easier to protect the receiving chain against unwanted oscillations .

[0035] The design described above is a typical solution for a total power radiometer and is af fected by its typical disadvantages : dependence on the temperature and operating conditions . The influence of these factors is too great to allow measurement of tissue radiation with accuracy suf ficient to determine glucose concentration without additional improvements .

[0036] The isolators can be replaced with microwave attenuators with a low attenuation, e . g . , 2 -3 dB . Such a solution is less ef fective and even increases the influence of chain noise but facilitates miniaturi zation .

[0037] In the present invention, the problem of the influence of the temperature and operating conditions is solved digitally . It was found that the glucose concentration in the tissue af fects both the noise generated by the tissue and the electrical parameters of the tissue , which af fects the reflectivity at the interface of the tissue and the antenna 2 , as well as the tissue temperature . During the measurement , the antenna reaches a temperature essentially equal to the temperature of the tissue surface . This latter circumstance adds disturbance in the form of the antenna ' s own noise . Additionally, the measurement performed with the detector 11 is disturbed by the noise of the receiving chain elements , which depends on their temperature and the current they consume . It turned out that the influence of the operating conditions ( the temperature and current ) of the receiving chain on the received noise signal power can be well modeled — j ust like the influence of blood glucose on the noise power . It can be done in one process .

[0038] To solve the problem of the influence of temperature changes on the gain of the radiometer chain and, further, the influence of these changes on the readings of the power detector11, contact temperature sensors (thermocouples) were provided in the following places of the receiving chain:- antenna — on the ground metalization side, patient temperature sensor 12,- amplifiers 3, 5, 7, 9 — on housings, temperature sensors 13,15, 18, 20, respectively .

[0039] Additionally, sensors 14, 16, 19, 21 for the supply current of individual amplifiers 3, 5, 7, 9, respectively, were provided .

[0040] The signal outputs of the sensors 12, 13, 15, 18, 20 14,16, 19, 21 and the output of the power detector are wired to the processing system 17, which contains an analog-to-digital converter and a microprocessor for processing measurement data. The power supply for the radiometer chain elements, the sensors, and the acquisition module is provided by a dedicated power module containing a lithium-ion battery and a set of voltage converters .

[0041] A person skill in the art is readily able to propose numerous alternatives to the processing system 17, including the use of microprocessors, digital signal processors (DSPs) , field programmable gate arrays (FPGAs) , and application specific integrated circuits (ASICs) for a dedicated signal processing task. Additionally, system-on-chip (SoC) solutions can be adopted, Raspberry Pi or single-board computers, Arduino-based microcontrollers, RISC-V processors, and ARM Cortex-M series microcontrollers have been tested to meet specific computing requirements. In particular, considerations include low-power microcontrollers for battery-powered applications. One can also use distributed solutions and send sensor and detector signals to remote cloud servers for processing, thereby increasing scalability, availability, and data management capabilities.

[0042] In this embodiment, the processing system 17 acquires and converts the output signals from the sensors 12, 13, 14, 15, 16, 18, 19, 20, 21 and the detector 11 into the glucose concentrationvalue using the multivariate regression method and predefined trans formation coef ficients . Key observation is that proposed set of coef ficients allows to discriminate glucose concentration from other variable parameters af fecting the total power of noise signal originating from tissue but measured with passive radiometer - even with the use of simple total power radiometer structure . This set covers first : the power of noise signal , the contribution to the power introduced by the impedance mismatch between the antenna and the tissue , the thermal noise contribution of antenna itsel f , the noise introduced by the active elements o f the receiver (mainly the shot-noise caused by bias current flow) , the changes of gain due to temperature of the active elements of the receiving chain of the radiometer . The noise introduced by the active elements correlates with their respective temperatures and currents flowing through the semiconductor structures (mainly semiconductor j unctions ) . The gain of the active elements correlates with their point of work and depends on the temperature . The temperature of active elements (measured with temperature sensors ) is required to model of the changes in their gain and the currents are required to model the noise contributed by the active elements . The impedance of tissue depends on glucose concentration . The noise power originating from the tissue is correlated with surface temperature of the skin covering the tissue ( surface temperature of the tissue ) combined with the deep tissue temperature .

[0043] The trans formation coef ficients are determined by performing a series of measurements of phantoms with known glucose concentration at various phantom and ambient temperatures while changing the operating conditions of the receiving chain . These coef ficients can also be determined by examining a group of patients in di f ferent conditions or even one patient for a long time while changing the operating conditions and regulating the glucose concentration and veri fying it in another measurement .

[0044] Alternatively, known arti ficial models corresponding to patients can be used as basis for determining coef ficients . A table approach is also acceptable , in which glycaemia values in tissue 1 are determined during reference measurements based on the adopted model .

[0045] Determination of glycemia should be preceded by reference measurements and determination o f model coef ficients . This can be done in a manner known in the art by measurements performed on simulated phantom tissues with di f ferent levels of glucose concentration or patients simultaneously subj ected to reference measurement by another method . Measurements performed on many patients or phantoms enable averaging and determining universal trans formation coef ficients . Measurement performed on a single patient can be used to obtain trans formation coef ficients dedicated to that speci fic person and potentially more accurate results .

[0046] It is important that the data set obtained from the sensors makes it possible to " extract" the influence of the glucose concentration from other factors af fecting the noise power and to determine the trans formation coef ficients , which enables the determination of glycemia in a measurement that includes not only the noise power, but also the temperature of the tissue surface , the temperatures of the receiving chain elements and the currents consumed by the active elements of the receiving chain . This can be done using multivariate regression or adopting a speci fic chain model or even using reference data to train a neural network running on the processing system 17 , receiving the data set from the sensors and detector and outputting the glycemia measurement result . In this context , the trans formation coef ficients are the weight coef f icients of the neural network .

[0047] The invention can even be implemented by creating a lookup table containing glucose concentration values corresponding tospeci fic sets of measured parameters . The entries of the table then constitute trans formation coef ficients .

[0048] The method according to the invention can be implemented in a general purpose computer fed with data or in a processing unit of the medical device . It should be stressed that person skilled in the art having learned the teaching of the present description, is able to routinely adapt multiple state of the art implement it in a digital device integrated with the radiometer or as well as run remotely e . g . on a remote server . The method according to the invention can be implemented in of the shel f digital devices e . g . computers or microcomputers or microcontrollers etc . Those skilled in the art given the teachings of the above description are able to routinely propose multiple hardware , software or hybrid solutions therefor .

[0049] The embodiments above are discussed with reference to a simple total power radiometer however the person skilled in the art will recogni ze that the invention can be used with any passive radiometers comprising total power radiometer including Dicke radiometer and correlation radiometer . Dike radiometer and correlation radiometer are more complex and more di f ficult to miniaturi ze but allow to reduce impact of the ambient temperature and conditions on the measurement result .

[0050] The computer program product according to the invention may be embodied on a tangible data carrier such as a CD, DVD, hard disk, or flash memory, recorded with data representing said computer program . Alternatively, the program can be stored " in cloud" or a non-tangible medium, such as a computer network connection .

[0051] Said computer program comprising instructions which, when executed by a computer, implement at least a portion of the method described in any one of the preceding embodiments . The computer program may include various elements such as subroutines , functions , procedures , obj ect methods , obj ect implementations , executable applications , applets , servlets ,source code , obj ect code , shared libraries , dynamic load libraries , and / or any other sequence of instructions intended for execution on a computer system .

[0052] It is important to emphasi ze that the embodiments mentioned above serve to illustrate , rather than limit , the scope of the invention . Those skilled in the art will readily recogni ze that various alternative embodiments may be designed without departing from the scope of protection defined by the claims .

[0053] In the claims , reference numerals should not be regarded as limiting the scope of the claims . The use of the verb " comprise" and its variations does not preclude the inclusion of additional elements or steps beyond those explicitly mentioned in a claim . Similarly, the indefinite articles "a" or " an" preceding an element do not rule out the presence of multiple such elements . The method according to the invention may be reali zed using computer hardware comprising multiple distinct components or by a single suitably programmed computer .Industrial applicability and advantages of the invention

[0054] As a result of using the solution according to the invention, additional advantages are obtained :- no need to use additional microwave elements such as input circulator and directional couplers , which facilitates miniaturi zation,- single chain instead of double chain,- no need to use additional reference noise sources , usually with the necessary temperature stabili zation,- no need to use an input switch to switch the input of the ampli fying chain successively to the antenna output and noise sources .

[0055] The device implementing method according to the invention according to the invention can be integrated in clothing and come into contact with the skin of the examined person . It is completely passive and does not emit any measurement signal to the patient ' s tissues . As a result , it is completely safe andthe measurement system and the method according to the invention can be used in the prevention and treatment of diabetes , especially in pregnant women and children under 2 years of age .

Claims

Claims1 . A computer implemented method of non-invasive determination of a blood glucose concentration in a tissue of a patient , the method being based on a radio noise signal received using an antenna kept in a proximity of a skin of the patient , characterized in that the radio noise signal is received with a passive radiometer and the method further comprises following steps : a step of obtaining trans formation coef ficients , a step of measuring a surface temperature of the tissue , a step of measuring temperatures of active elements of a receiving chain of the radiometer, a step of measuring currents consumed by the active elements of the receiving chain of the radiometer, a step of measuring a power of the radio noise signal originating from the tissue , a step of determining the blood glucose concentration in the tissue based on said trans formation coef ficients , said measured values of the temperature of the tissue surface , said temperatures of active elements of the receiving chain of the radiometer said currents consumed by the active elements o f the receiving chain of the radiometer, and the power of the radio noise signal originating from the tissue .2 . Method according to claim 1 , which is implemented in a computer, and in the step of obtaining trans formation coef ficients , predefined trans formation coef ficients are loaded from a computer memory or downloaded from an external server .3 . Method according to claim 1 , in which the transformation coef ficients are determined based on a series of measurements performed on a set of tissues or phantoms of known glucose concentrations , the measurements are being performed at di f ferent temperatures .4 . Method according to claim 3 , in which the transformation coef ficients are determined in the process of multivariate regression consi sting in minimi zing the error of matching glucose concentration reference values to the values calculated from the transient characteristics of the radiometer, wherein the variables are : the temperature of the tissue surface , the temperatures of the active elements of the receiving chain of the radiometer, the currents consumed by the active elements of the receiving chain of the radiometer, and the power of the radio noise signal .5 . Method according to any of claims from 1 to 4 , in which converting the measured values and the trans formation coef ficients into the blood glucose concentration includes the following steps : determining the power of the noise received by the antenna based on the trans formation coef ficients , the power measured using the radiometer, the temperatures of the active elements of the receiving chain of the radiometer , and the currents consumed by the active elements of the receiving chain of the radiometer, determining the power of the noise originating from the tissue and reaching the antenna and reflection coef ficient at the interface between the antenna and the tissue based on the power of the noise received by the antenna, the trans formation coef ficients , and the temperature of the surface of the tissue ,determining the glucose concentration based on the determined reflection coef ficient at the interface of the antenna and the tissue and the determined power of the noise originating from the tissue and reaching the antenna .6 . Method according to claim 1 or 2 or 3 , in which the trans formation coef ficients are weight coef ficients of a neural network processing the data set containing the measured temperature of the tissue surface , the temperatures of active elements of the receiving chain of the radiometer the currents consumed by the active elements of the receiving chain of the radiometer, and the power of the radio noise signal originating from the tissue .7 . Method according to any of claims from 1 to 6 , in which the measurements of the radio noise signal are carried out in the band ranging from 800 MHz to 24 GHz .8 . Method according to any of claims from 1 to 7 , in which the measurements are carried out in a band with a width greater than or equal to 200 MHz .9 . Method according to claim 8 , in which the measurements are carried out in a band with a width greater than or equal to 500 MHz .10 . A computer program product to be run on a computer adapted to receive measurement data characterized in that it comprises instructions that causes the method defined in any of claims from 1 to 9 to be performed .

11. A radiometer comprising an antenna (2) and a radio receiving chain (3, 4, 5, 6, 7, 8, 9, 10) with a detector (11) , the antenna (2) being adapted to be applied to a skin (1) of a patient and to receive a radio noise signal from a subcutaneous tissue, characterized in that the radiometer is a passive radiometer adapted to measure power of noise signal originating from the tissue equipped with sensors (13, 15, 18, 20) of temperatures of the active elements of the receiving chain of the radiometer, a sensor (12) of a temperature of the tissue surface located on the antenna (2) , and sensors (14, 16, 19, 21) of the currents consumed by the active elements of the receiving chain of the radiometer.

12. The radiometer according to claim 11, in which the antenna has a frequency band at least 20% wider than the frequency band of the receiving chain.

13. The radiometer according to claim 11, in which the receiving chain contains a cascade of amplifiers (3, 5, 7, 9) separated by isolators (4, 6, 8) .

14. The radiometer according to claim 11, in which the receiving chain contains a cascade of amplifiers (3, 5, 7, 9) separated by attenuators with a rated attenuation below 3 dB .

15. A system for determining glucose concentration based on a radio signal, comprising a processing system (17) and a radiometer with an antenna (2) adapted to be applied to a tissue, characterized in that the radiometer is radiometer according to any of claims from 11 to 14, and the processing system (17) has connected theretosensors (13, 15, 18, 20) of the temperatures of the active elements of the receiving chain of the radiometer, a sensor (12) of the temperature of the tissue surface, placed on the antenna (2) , current measurement modules (14, 16, 19, 21) measuring the currents consumed by the active elements of the receiving chain of the radiometer, and the processing system (17) is adapted to carry out the method as defined in any of claims from 1 to 9.

Images