Non-invasive measurement of biomarker concentration

Abstract

The present invention relates to a device (100) for determining a biomarker concentration in the blood of a body part (110) taking into account the physiological composition of the body part (110). The device (100) comprises a light source (101) for irradiating the body part (110) with a first light wave (104), a detector unit (102) for measuring the first light wave (104) reflected from the body part (110), and a processing unit (103) coupled to the detector unit (102) for receiving the measured first light wave (104). The processing unit (103) is configured to determine at least one characteristic value including a signal strength (SS) of the reflected first light wave (104) upon the occurrence of a first specific signal section (202) in the signal profile (201) of the reflected first light wave (104) during a predetermined pressure variation applied to the body part (110) by the detector unit (102), wherein the at least one characteristic value in the specific first signal section (202) of the reflected first light wave (104) represents a physiological composition of the body part (110) and enables a biomarker concentration in the blood to be determined.

Description

Technical Field

[0001] The present invention relates to a device for determining the concentration of biomarkers in the blood of a body part such as a finger, taking into account the physiological composition of the body part. Further, the present invention relates to a method for determining the concentration of biomarkers in the blood of a body part such as a finger, taking into account the physiological composition of the body part.

Background Art

[0002] For measuring biomarker concentrations, such as blood glucose concentration, invasive measurement methods and respective measurement devices exist. To determine each biomarker concentration, blood is drawn from a person's tissue and each blood is analyzed.

[0003] Furthermore, non-invasive measurement methods are known. For example, there are devices that irradiate a person's tissue with respective light such as infrared light having a specified wavelength. Based on the measured reflected light, it is generally possible to determine the generation and concentration of a specific biomarker. However, conventional non-invasive measurement methods are not very accurate due to the diverse physiological compositions of body parts and many other environmental measurement parameters.

[0004] One reason for inaccurate measurement results is that the physiological functions of the measured body parts such as fingers vary and change very rapidly over time. The physiological functions of the measurement site can be defined, for example, by finger temperature, skin thickness, blood circulation in the subcutaneous tissue, subcutaneous thickness, bone depth, skin color, and skin moisture, for example.

[0005] Furthermore, in conventional measurement methods, the pressure of body parts such as fingers on each detection device being indefinite can lead to inaccurate measurement of each biomarker concentration.

[0006] For example, International Publication No. 2016 / 068589 discloses a glucose measurement device that measures blood glucose levels based on infrared spectroscopy. A pressure sensor is used to measure the pressure applied from the body part to the device in order to determine the measurement error caused by the unknown pressure between the body part and the detection unit.

[0007] International Publication No. 00 / 21437 discloses an infrared glucose measurement system that uses attenuated total internal reflection spectroscopy. This measurement system includes a pressure maintaining member to maintain a predetermined pressure between the body part and each detection plate of the measurement system.

[0008] Therefore, any non-invasive measurement device is inaccurate, so that reliable measurement results are not possible, or a complex device needs to be provided to pre-determine a specific pressure.

Summary of the Invention

[0009] Therefore, an object of the present invention is to provide a simple measurement device that further provides a highly accurate non-invasive measurement of biomarkers in human blood.

[0010] This object is solved by a device and a method for determining the concentration of a biomarker in the blood of a body part, taking into account the physiological constitution of the body part, according to the subject matter of the independent claims.

[0011] According to a first aspect of the present invention, there is provided a device for determining the concentration of a biomarker in the blood of a body part, such as a human finger, taking into account the physiological constitution of the body part. The device includes a light source that irradiates a first light wave to the body part, and a detector unit that measures the first light wave reflected from the body part. Further, the device includes a processing unit coupled to the detector unit that receives the measured first light wave.

[0012] The processing unit is configured to determine at least one characteristic value including the signal intensity of the reflected first light wave when a first specific signal section in the signal profile of the reflected first light wave during a predetermined pressure fluctuation applied to the body part by the detector unit occurs. At least one characteristic value in the specific first signal section of the reflected first light wave represents the physiological constitution of the body part, and the concentration of the biomarker in the blood can be determined.

[0013] According to a further aspect of the present invention, a method for determining the concentration of a biomarker in the blood of a body part is presented in consideration of the physiological constitution of the body part. The method includes the steps of irradiating the body part with a first light wave, measuring the reflected first light wave from the body part, and determining at least one characteristic value including the signal intensity of the reflected first light wave when a first specific signal section in the signal profile of the reflected first light wave during a predetermined pressure fluctuation applied to the body part by the detector unit occurs. At least one characteristic value in the specific first signal section of the reflected first light wave represents the physiological constitution of the body part, and the concentration of the biomarker in the blood can be determined.

[0014] The device can be a portable device, in particular, a smartphone, a tablet computer, or a notebook.

[0015] The determined biomarker can be glucose, C-reactive protein (CRP), hemoglobin (HBC), cholesterol, LDL, HDL, fibrinogen, and / or bilirubin.

[0016] The light source is configured to irradiate a body part with light having a first wavelength or a plurality of additional predetermined wavelengths. The light source may include one or more LEDs. Specifically, the first wavelength may have, for example, 420 nm to 490 nm (blue light), 490 nm to 575 nm, particularly 530 nm (green light), 585 nm to 750 nm, particularly 660 nm (red light), and 780 nm and 1000 nm, particularly 960 nm (infrared IR light).

[0017] The detector unit may have a photodiode configured to measure all relative described spectra used for each irradiation wavelength. Specifically, the detector unit may detect a photograph or a plurality of spectra, for example, from 410 nm to 1090 nm.

[0018] The detector unit may measure the illuminance of the received reflected wavelength in units of [lux]. Next, in the signal acquisition process, the measured illuminance is converted into, for example, a Row-ADC signal having a unit of [nA] (nanoampere). The value of the signal intensity in nA units may be, for example, from 0 to 224000 nA. However, this value depends on the sensor (detector unit) used and thus may vary when different sensors are used.

[0019] The processing unit may have a processor that controls the light source and the detector unit. Specifically, the processing unit may have, for example, an oscillator, an LED driver, a temperature sensor, and a data register. Further, it processes the transfer of data via a standard bus such as I2C or SPI communication.

[0020] Furthermore, the device may include a display unit that displays the measurement results and / or gives instructions to the user. Further, the display unit may form an input unit such as a touch screen.

[0021] The quality and quantity of the signal intensity of the wavelength that is reflected and thus detected depend on the physiological constitution of the body part, and specifically on the pressure when the detection unit is pressed against the body part. With the approach of the present invention, it has been found that, independent of the measured pressure value applied to the body part and the knowledge of the physiological constitution of the body part, the detected signal during a predetermined pressure variation can represent the amount of biomarker concentration.

[0022] The pressure variation can be, for example, an increase and decrease in pressure at regular time intervals. The pressure variation can be independent of the initial pressure and the final pressure of the pressure variation. For example, (one) predetermined pressure variation can be an increase and decrease in pressure within a time span of, for example, 10 to 20 seconds.

[0023] In the signal profile of the detected reflected light wave during a given pressure variation, it has been found that there are signal intervals (for example, a certain shape) specific during a given pressure irradiation. Furthermore, it has been found that the specific signal intervals and their respective characteristic values (for example, the intensity of the detected signal in the specific signal interval) indicate a certain biomarker (for example, glucose) and its respective concentration. Furthermore, it has been found that the characteristic values derived from the signals of the specific signal intervals can define the specific physiological constitution of the body part at the time of measurement. For example, when the body part is a finger and the finger is pressed against the detection unit during a given pressure variation, the maximum value as the signal interval of the detected signal profile can indicate the amount of tissue between the surface of the finger and the bone of the finger. Therefore, the thickness of the tissue between the bone and the surface of the finger can be derived, which also affects the measurement result of the biomarker concentration.

[0024] Each specific point and specific signal interval in the signal profile can be a flat region of the signal function, a steep change part of the function (a sudden change in the slope of the function), and the maximum and minimum values of the signal function.

[0025] Therefore, since pressure fluctuations with an undetermined initial pressure can be performed by the user without measuring the total amount of pressure at a certain point in time, a complex pressure sensor is not necessary in the present invention. Furthermore, by determining the characteristic values of specific signal intervals during a given pressure fluctuation, it leads to a more accurate determination of biomarker concentration and a more accurate measurement system is provided.

[0026] The determined characteristic values in the specific signal intervals of the reflected light waves can be compared with existing models that contain information on the respective biomarker concentrations in the blood at certain specific characteristic values of the specific signal intervals. The existing models are defined, for example, in clinical and laboratory studies. For example, if the biomarker is glucose, the glucose values and physiological constitutions of multiple people can be measured invasively, for example. For example, an oral glucose tolerance test (OGTT) can measure accurate glucose values for a user's specific physiological constitution. For the measured glucose values, the specific characteristic values of the specific signal intervals in the signal profile of the reflected light waves can be determined. Therefore, a database can be provided that contains a plurality of nominal values that can be compared with the measured characteristic values of the device of the present invention to determine the concentration of a specific biomarker in the blood. In fact, a plurality of specific signal intervals in a plurality of different light waves can be derived for a specific biomarker concentration taking into account the specific physiological constitution. For example, as described below, statistical research methods based on respective defined regressors and regressor relationships can be used to further increase the accuracy of the determined concentration levels of the biomarker.

[0027] According to a further exemplary embodiment, the characteristic value further has the value of the slope of the signal profile at the occurrence of a specific signal interval during the given pressure fluctuation applied to the body part by the detector unit.

[0028] According to a further exemplary embodiment, the specific signal interval is defined by characteristic slopes, by the flat region of the signal function, by the sharp turn of the signal function, by inflection points, by minimum values, in particular the minimum value of the signal function, and by maximum values, in particular the maximum value of the signal function. Thus, during a given pressure fluctuation, each signal profile of the reflected light wave includes, for example, the specific signal intervals mentioned above that indicate the biomarker concentration and physiological constitution of a body part.

[0029] According to a further exemplary embodiment, the processing unit is configured to determine, for each executed pressure fluctuation, based on a plurality of repeated given pressure fluctuations occurring in the signal profile of the first reflected light wave. The processing unit is further configured to determine each characteristic value of the first specific signal interval in each given pressure fluctuation and to determine the average characteristic value of the first specific signal interval determined in the given pressure fluctuation. Thus, if the pressure fluctuation is an increase in pressure for 10 seconds and the user is only increasing the pressure for 5 seconds, an error measurement may occur. However, by providing multiple measurements during multiple pressure fluctuations, the influence of one error measurement is reduced by the average value of all measurements.

[0030] According to a further exemplary embodiment, at least one determined characteristic value defines at least one respective characteristic regressor (Rc). The processing unit is configured to determine a regressor relationship (RR) based on at least one determined characteristic regressor (Rc), and the regressor relationship is correlated with the biomarker concentration in the blood, and the determined value of the regressor relationship is adapted to indicate the value of the biomarker concentration.

[0031] The characteristic values at the occurrence of this specific signal interval define a first list of regressors, i.e., characteristic regressors (Rc). This list of regressors Rc can be used to generate a regressor relationship that may be correlated with the biomarker concentration. The regressor relationship defines a mathematical relationship between at least one characteristic regressor or the relationships of a plurality of different characteristic regressors. By using statistical methods and machine learning, i.e., artificial intelligence (AI), based on the characteristic values, a specific regressor relationship suitable for determining the biomarker concentration of a specific physiological constitution can be obtained based on the specific signal interval of the signal profile of the light wave reflected under a given pressure fluctuation.

[0032] The term "machine learning" may in particular mean the implementation of algorithms and / or statistical models that a processor (such as a computer system) can use to obtain each regressor relationship that best matches the biomarker concentration in a specific physiological constitution of a body part. Through machine learning, the best-matching regressor relationship can be obtained by relying on patterns and estimations instead of using explicit instructions. Machine learning can be regarded as a subset of artificial intelligence. In particular, machine learning algorithms can build a mathematical model based on the regressor relationship regarding sample data (biomarker concentrations measured under laboratory conditions, i.e., invasive, or in the case of using glucose as a biomarker, such as those from the above-mentioned OGTT test) to make predictions or decisions without explicitly programming for the execution of tasks. Machine learning algorithms can be particularly appropriately applied in the evaluation of regressors and the regressor relationship indicating the specific signal interval in the signal profile of the reflected wavelength.

[0033] In one embodiment, machine learning uses at least one of the group consisting of random forest, random fern, support vector machine, and neural network, particularly convolutional neural network.

[0034] The term "random forest" can mean an ensemble learning method for classification, regression, and other tasks that operates by, in particular, constructing a large number of decision trees during training time and outputting the class (which can mean classification) or the average prediction (which can mean regression) that is the mode of the classes of the individual trees.

[0035] The term "random fern" can mean a machine learning algorithm that, in particular, enables the recognition or tracing of an object (such as a solid pharmaceutical composition or a part thereof) by matching the same elements between two images of the same scene. Random fern can be implemented as a classification method.

[0036] The term "support vector machine" can mean a supervised learning model associated with a learning algorithm that analyzes data used for classification and regression analysis. Given a set of training examples each marked as belonging to one or the other of two categories, the support vector machine training algorithm can construct a model that assigns new examples to one or the other category. A support vector machine model can represent examples as points in a space that are mapped such that examples of separate categories are divided by the widest possible clear gap. New examples can then be mapped into the same space and it can be predicted that they belong to the category based on the side of the gap to which they correspond.

[0037] The term "neural network" (or artificial neural network) generally refers to a computing system that can learn to perform a task by considering examples, without specifically programming task-specific rules, which can be inspired by the biological neural circuitry that makes up the human or animal brain. A neural network can identify patterns without prior knowledge of the object to be identified (e.g., the coating of a solid composition). Additionally or alternatively, a neural network can automatically generate discriminative features from examples of training data that the neural network processes. A neural network can be based on a collection of connected units or nodes that can represent artificial neurons. Each connection between different nodes can transmit a signal to other neurons. Then, an artificial neuron that receives the signal can process it and transmit the signal to the neurons connected to it.

[0038] Accordingly, machine learning can be implemented in an evaluation to find an appropriate regressor relationship. Detection data of the reflected wavelengths obtained under laboratory conditions by a detection unit can be at least partially analyzed using machine learning. This can make it possible to obtain highly reliable information regarding a regressor relationship indicating the biomarker concentration in a specific physiological constitution of a specific body part (such as a finger).

[0039] According to the present invention, such a composition (e.g., the regressor relationship of regressors in several signal intervals in the signal profiles of different wavelengths) can show a reliable prediction of the biomarker concentration. Therefore, it has been found that the regressor relationship indicating a specific signal interval in the signal profile of the reflected wavelength is suitable for evaluation by machine learning.

[0040] An appropriate regression relationship that determines the biomarker concentration of a specific biomarker in a specific physiological constitution of a specific body part (e.g., a human finger, lip, etc.) is correlated with the laboratory measurement results of the human biomarker concentration in laboratory tests and can be stored in respective databases. Therefore, when measuring a specific characteristic value for a specific regression relationship, since the influence of the actual biomarker concentration has already been considered by the regression relationship, the respective concentrations of the biomarkers can be determined by comparing them with the respective nominal values of the regression relationships in the database without determining the user's physiological constitution.

[0041] According to a further exemplary embodiment, the device further comprises a data unit having a dataset of a predefined regression relationship correlated with each biomarker concentration. The control unit is further configured to compare the determined regression relationship with the predefined regression relationship, and when the determined regression relationship is in the vicinity of the predefined regression relationship, the biomarker concentration is derivable.

[0042] The data unit can be implemented within the device. However, the data unit can be realized by the input / output interface of the device, and data can be received and / or transmitted between the device and a remote data unit that stores the data. Therefore, a web-based application can be used, the data can be stored in a (web) server or a cloud server, and the device can receive and / or transmit data via the Internet or other network connections.

[0043] According to a further exemplary embodiment, the processing unit is further configured to determine at least one further characteristic value including the further signal intensity of the reflected first light wave when a further first specific signal section in the signal profile of the reflected first light wave during a predefined pressure variation applied to the body part by the detector unit occurs, and at least one further characteristic value in the further specific first signal section of the reflected first light wave represents the physiological constitution of the body part and enables the determination of the biomarker concentration in the blood. At least one determined further characteristic value of the further specific first signal section defines at least one respective further characteristic regressor (Rcf), and the regressor relationship is further determined based on at least one determined further characteristic regressor (Rcf). According to an exemplary embodiment, it is outlined that the signal profile of the reflected wavelength may have a plurality of specific signal sections that can be used as regressors defining the regressor relationship. Thus, the regressor relationship (RR) is formed by a mathematical dependency and the relationships of the characteristic regressor (Rc) and the further characteristic regressor (Rcf).

[0044] According to a further exemplary embodiment, the processing unit is further configured to determine at least one measurement value of the signal intensity of the reflected first light wave during a particularly constant arrangement of the detector unit with respect to the body part (and thus a substantially constant pressure), and the measurement value defines at least one measurement regressor (Rm). The regressor relationship (RR) is further determined based on at least one determined characteristic regressor (Rc, Rcf) and at least one measurement regressor (Rm).

[0045] During the placement of the detector on the body part, i.e., under (substantially) constant pressure, by further measuring the reflected wavelength, it has been found that the characteristic value of the reflected signal of a specific wavelength obtained under substantially constant pressure can define a measurement regressor that can be used for the normalization of data regarding the current physiological constitution of the body part and regarding the calibration of a light source, for example an LED. Since the measurement regressor is further considered in the regressor relationship, an improved criterion for the regressor relationship with respect to the nominal regressor relationship indicating the biomarker concentration can be obtained.

[0046] According to a further exemplary embodiment, the light source is configured to irradiate the body part with a second light wave, and the detector unit is configured to measure the second light wave reflected from the body part. The detector unit is configured to receive the second light wave to be measured. The processing unit is further configured to determine at least one further characteristic value including the signal intensity of the reflected second light wave when a second specific signal section in the second signal profile of the reflected second light wave during a predefined pressure variation applied to the body part by the detector unit occurs, and at least one further characteristic value in the specific second signal section of the reflected second light wave represents the physiological constitution of the body part. At least one determined further characteristic value defines at least one respective further characteristic regressor, and the regressor relationship is further determined based on at least one determined further characteristic regressor (Rc2).

[0047] It is outlined by the above exemplary embodiments that specific spectra of different wavelengths can be irradiated and received by the device, and the regressor relationship is further formed by further characteristic regressors indicating signal sections of the signal profiles of further different wavelengths.

[0048] In summary, during the measurement of the reflected light of a body part under a given pressure fluctuation in the body part, each reflected wavelength (e.g., red light, infrared light, blue light, green light, etc.) has a specific signal profile under a specific pressure fluctuation applied to the body part. It has been found that each signal profile under pressure fluctuation has a respective specific signal interval in the signal profile of the first light wave reflected, which indicates the physiological constitution and / or the concentration of the measured biomarker.

[0049] At least one characteristic value including the signal intensity at the occurrence of the first specific signal interval during a given pressure fluctuation applied to the body part by the photosensor can be derived. The characteristic value from this signal profile may include the signal intensity in the specific signal interval and / or the value of the slope (differential) at the specific point.

[0050] According to a second finding of the present invention, it has been found that many specific regressor relationships of regressors can correlate much better with the biomarker concentration in the blood (e.g., glucose value). As the mathematical relationship between the characteristic regressor (Rc) and, for example, the measured regressor (Rm), the following specific regressor relationships can be obtained: Rm1 / Rc1, Rm2 / Rc1, Rm1 / Rc2, Rm1 / ln(Rc1), ln(Rm1) / e Rc1 etc.

[0051] The regressor Rm (by the second part of the measurement under a substantially constant pressure) does not correlate well enough with the biomarker in the blood due to the lack of information on the specific physiological function of the skin during the measurement.

[0052] By combining the regressor Rm with the regressor Rc, a specific regressor relationship RR can be generated, which, for example, forms the input regressor (Ri). The input regressor Ri correlates much better with the biomarker concentration in the blood (e.g., glucose value). The procedure of mathematical correction of the measured regressor by the characteristic regressor can be called physiological normalization.

[0053] Accordingly, the combination of the above important findings results in a very accurate non-invasive measurement of the concentration of biomarkers such as glucose in the blood of each body part. Specifically, through physiological normalization by measurement under a given pressure fluctuation, the measurement is normalized, so that the physiological constitution of the body part at the time of measurement no longer dramatically affects the quality of the measurement result at the time of measurement. Furthermore, by using a specific regressor relationship through measurement under a substantially constant pressure, a very accurate correlation to the desired biomarker concentration is possible.

[0054] In fact, each wavelength (infrared, green, red, etc.) defines a specific signal profile and its respective specific signal profile interval under a given pressure fluctuation (the first part of the measurement). Accordingly, based on a number of signal profiles and their respective specific signal intervals, multiple regressors can form a more complex specific regressor relationship. Such a complex specific regressor relationship for a specific biomarker concentration can be formed, for example, by applying a mathematical / statistical algorithm. Such a complex specific regressor relationship can be very successfully used as an input parameter (regressor) for regression analysis using machine learning and artificial intelligence.

[0055] In an exemplary measurement procedure performed by the device of the present invention, a first list of regressors Rc is received by pressing a photodetector against a body part (e.g., on a finger) during a given pressure interval (e.g., from a minimum pressure to a maximum pressure), and a second list of regressors Rm is received by positioning the photodetector relative to the finger to provide a substantially constant pressure.

[0056] Furthermore, the measurement cycles at various pressures and at a constant pressure can be repeated multiple times to provide an appropriate average value of the regressors to improve the quality of the measured values. Furthermore, prior to performing the measurement, respective calibrations of the light emitting element and each photodetector can be performed.

[0057] The device can be a smartphone or can also function as a stand-alone device with appropriately added components such as a processor, a screen, power management, a communication module, a battery, a charger, etc. When the measurement is carried out, the skin can be in direct contact with the surface. When the measurement starts, the sensor first turns on a light source, for example, a photodiode, and measures the current on the light source. In this way, the sensor can solve the problem of the torque of the current generated in the light source itself due to ambient light, environmental influence or physical parameters of the light source. In that case, the device drives the diodes of the light source individually, for example, at a frequency of 20 Hz to 100 Hz.

[0058] When the device is covered by a body part (for example, the pulp of a finger, preferably the index finger or the ring finger), a fixing element of the device (a rubber ring, a rubber string or any other elastomer, a rope, a fastener, etc.) can be used to fix the body part to the device for more accurate measurement.

[0059] Next, the individual measurement is started. Each person has different skin types and other physiological characteristics that can be evaluated by the device of the present invention. According to the present invention, the skin surface is pushed against the device with, for example, three consecutive pressures, whereby blood is squeezed out from the body part (for example, the tip of the finger) and the body part shrinks slightly. For example, first, the pressure is gradually increased until the diode signal is no longer distinguishable, which lasts for about 10 seconds, for example, then the pressure is gradually released for 5 seconds, for example, and then the entire process can be repeated, for example, more than 2 times. In that case, the body part can remain on the device for about 20 seconds under a substantially constant pressure.

[0060] Next, first, the validity and quality of the signal can be checked. Next, according to the present invention, the physiological functions of the skin of the finger can be examined based on the relationship between the above-described regressor and the regressor relationship RR. The physiological functions of the body part are examined in each regressor relationship. Based on the data of such measurements, it is possible to determine the actual skin and subcutaneous characteristics based on the first part of the measurement under pressure fluctuations and perform physiological normalization (FN) for the second part of the measurement under a substantially constant pressure. Physiological normalization is used to normalize the data of the second part of the measurement by converting the values into a neural (universal) model (database), where all the obtained values have, for example, the same scale (unit). Based on the data of the regressor relationship, it is possible to determine the position of the actually measured regressor relationship within the multidimensional space of the database of the nominal regressor relationship that correlates with the concentration of the biomarker, for example, the blood glucose level. The position of the measured regressor relationship within the multidimensional space of the database is determined, for example, based on clustering that determines the position of the statistical model within the spectral space of the model based on the data of the first part of the measurement. A more detailed classification of the measured regressor relationship can be provided by checking the relationship between signals of different wavelengths within a given measurement range.

[0061] According to yet another exemplary embodiment of the present invention, program elements (e.g., software routines in source code or executable code) are provided, which are adapted to control or execute a method having the above-described features when executed by a processor, e.g., a processor unit (microprocessor, CPU, GPU, FPGA, or ASCI, etc.).

[0062] According to yet another exemplary embodiment of the present invention, a computer-readable medium (e.g., CD, DVD, USB stick, floppy disk, hard disk, flash drive, or Blu-ray (registered trademark) disk) is provided, on which a computer program is stored, which is adapted to control or execute a method having the above-described features when executed by a processor (such as a microprocessor, CPU, GPU, FPGA, or ASCI, etc.).

[0063] Data processing that can be performed in accordance with embodiments of the present invention can be realized by a computer program (e.g., an application (app) installed on a smartphone), i.e., by software, or by using one or more special electronic optimization circuits, i.e., in hardware or in a hybrid form, i.e., by software components and hardware components.

[0064] It should be noted that embodiments of the present invention are described with reference to different subjects. In particular, some embodiments are described with reference to claims related to apparatuses, while other embodiments are described with reference to claims related to methods. However, those skilled in the art will appreciate from the above and the following description that, unless otherwise noted, any combination of features belonging to one type of subject, in addition to any combination between features related to different subjects, particularly between features of claims related to apparatuses and features of claims related to methods, is also considered to be disclosed in this application.

Brief Description of the Drawings

[0065] The above-defined and further aspects of the present invention will be apparent from the examples of embodiments described below and will be described with reference to the examples of embodiments. The present invention will be described in more detail below with reference to the examples of embodiments, but the present invention is not limited thereto.

[0066]

Figure 1

[0067]

Figure 2

[0068]

Figure 3

DETAILED DESCRIPTION OF THE INVENTION

[0069] Note that the illustrations in the drawings are schematic diagrams. The same or identical elements in different drawings are provided with the same reference numerals.

[0070] FIG. 1 shows a schematic diagram of a device according to an exemplary embodiment of the present invention. FIG. 2 shows a figure showing detected signals of different wavelengths under pressure fluctuations by the device according to FIG. 1.

[0071] The illustrated device 100, such as a smartphone, determines the biomarker concentration in the blood of a body part 110, such as the fingertip shown, taking into account the physiological constitution of the body part 110. The device 100 includes a light source 101 that irradiates the body part 110 with a first light wave 104, a detector unit 102 that measures the first light wave 104 reflected from the body part 110, and a processing unit 103 coupled to the detector unit 102 that receives the measured first light wave 104. When a first specific signal section 202 in the signal profile 201 of the reflected first light wave 104 occurs during a predetermined pressure variation applied to the body part 110 by the detector unit 102, the processing unit 103 is configured to determine at least one characteristic value including the signal intensity SS of the reflected first light wave 104. At least one characteristic value in the specific first signal section 202 of the reflected first light wave 104 represents the physiological constitution of the body part 110, and the biomarker concentration in the blood can be determined.

[0072] The light source 101 is configured to irradiate the body part with light having a first wavelength 104 or a plurality of additional predetermined wavelengths 204, 205. The light source 101 may include one or more LEDs. Specifically, the first wavelength 104 may be infrared light, the second wavelength 204 may be blue light, and the third wavelength 205 may be green light.

[0073] The detector unit 102 may have a photodiode configured to measure all relative spectra used for each irradiation wavelength 104, 204, 205. Specifically, the detector unit 102 may detect a photograph or a plurality of spectra, for example, from 410 nm to 1090 nm.

[0074] The processing unit 103 may have a processor that controls the light source 101 and the detector unit 102. Specifically, the processing unit may have, for example, an oscillator, an LED driver, a temperature sensor, and a data register (for example, the data unit 105). Further, it performs processing for transferring data via a standard bus such as I2C or SPI communication.

[0075] Furthermore, the device 100 may include a display unit 106 for displaying the measurement results and / or giving instructions to the user. Further, the display unit 106 may form an input unit such as a touch screen.

[0076] The quality and amount of the signal intensities of the wavelengths 104, 204, 205 that are reflected and thus detected depend on the physiological constitution of the body part 110, and specifically on the pressure when the detection unit 100 is pressed against the body part 110. However, independently of the pressure applied to the body part 110 and the physiological constitution of the body part 110, the detected signal during a predetermined pressure variation can represent the amount of biomarker concentration.

[0077] During a predetermined pressure variation in the signal profiles 201, 206 of the detected reflected light waves, specific signal intervals 202, 207 (for example, a certain shape) exist during the predetermined pressure irradiation. Further, it has been found that the specific signal intervals 202, 207 and their respective characteristic values (for example, the intensity of the detected signal in the specific signal intervals 202, 207) indicate a specific biomarker (for example, glucose) and its respective concentration. The value of the signal intensity SS can be 0 to 224000 nA in the example shown in FIG. 2. In FIG. 2, the pressure variation, and thus the signal intensity variation, over time t for the wavelengths 104, 204, 205 is shown.

[0078] Furthermore, it has been found that the characteristic values derived from the signals of the specific signal intervals 202, 207 can define the specific physiological constitution of the body part at the time of measurement. For example, when the body part 110 is a finger and the finger is pressed against the detection unit 102 during a predetermined pressure variation, the maximum values as the signal intervals 202, 207 of the detected signal profiles 201, 206 can indicate the amount of tissue between the surface of the finger and the bone of the finger. Accordingly, the thickness of the tissue between the bone and the surface of the finger can be derived, which also affects the measurement result of the biomarker concentration.

[0079] The respective specific points and specific signal intervals 202, 207 in the signal profiles 201, 206 can be flat regions of the signal function, sudden change portions of the function (sudden changes in the slope of the function), and the maximum and minimum values of the signal function.

[0080] The determined characteristic values in the specific signal intervals 202, 207 of the reflected light waves 104, 204, 205 can be compared with an existing model that includes information on the respective biomarker concentrations in blood at a certain characteristic value of the specific signal intervals 202, 207. The existing model is defined, for example, in clinical studies and laboratory studies. For example, when the biomarker is glucose, the glucose values and physiological constitutions of multiple people can be measured invasively, for example. For example, an oral glucose tolerance test (OGTT) can be used to measure accurate glucose values regarding a user's specific physiological constitution. For the measured values of the glucose values, the specific characteristic values of the specific signal intervals 202, 207 in the signal profiles 201, 206 of the reflected light waves can be determined. Therefore, a database stored, for example, in the data unit 105 and including a plurality of nominal values can be provided and compared with the measured characteristic values of the device of the present invention to determine the specific biomarker concentration in blood. In fact, the plurality of specific signal intervals 202, 203, 207 in a plurality of different light waves can be derived for a specific biomarker concentration in consideration of a specific physiological constitution.

[0081] The specific signal interval 202 shows, for example, a maximum value. A further specific first signal interval 203 shows, for example, an inflection point. The second signal interval 207 of the second signal profile 206 shows, for example, a flat region of the signal function. Therefore, during a given pressure fluctuation, each signal profile 201, 206 of the reflected light waves includes the specific signal intervals 202, 203, 207 mentioned above, for example, indicating the biomarker concentration and physiological constitution of the body part 110.

[0082] The processing unit 103 is configured to determine, for each of the executed pressure fluctuations, based on a plurality of repeated predefined pressure fluctuations occurring in the first specific signal sections 202, 203, 207 in the signal profiles 201, 206 of the reflected first light waves. The processing unit 103 is further configured to determine the characteristic value of each of the first specific signal sections in each predefined pressure fluctuation, and to determine the average characteristic value of the first specific signal sections 202, 203, 207 determined in the predefined pressure fluctuations.

[0083] At least one determined characteristic value of the specific signal sections 202, 203, 207, for example, signal intensity or signal slope, defines at least one respective characteristic regressor (Rc, Rcf). For example, the signal profiles 201, 206 of the reflected wavelengths 104, 204, 205 may have a plurality of specific signal sections 202, 203, 207 that can be used as regressors defining a regressor relationship. Thus, the regressor relationship RR is formed by a mathematical dependency and the relationship of the characteristic regressor (Rc) and the further characteristic regressor Rcf.

[0084] The data unit 105 of the device includes a dataset of a predefined regressor relationship RR that correlates with each biomarker concentration. The processing unit 103 is further configured to compare the determined regressor relationship RR with the predefined regressor relationship RR, and when the determined regressor relationship RR is in the vicinity of the predefined regressor relationship, the biomarker concentration is derivable.

[0085] The characteristic values at the occurrence of these specific signal intervals 202, 203, 207 define a first list of characteristic regressors Rc, Rcf. This list of regressors Rc, Rcf can be used to generate a regressor relationship RR that may be correlated with biomarker concentration. The regressor relationship RR defines a mathematical relationship between at least one characteristic regressor Rc, Rcf, or the relationship of a plurality of different characteristic regressors. By using statistical methods and machine learning, i.e., the use of artificial intelligence (AI), a specific regressor relationship suitable for determining the biomarker concentration of a specific physiological constitution based on the specific signal intervals 202, 207 of the signal profiles 201, 206 of the light waves reflected under a given pressure fluctuation can be obtained from the characteristic values.

[0086] Therefore, machine learning can be implemented in the evaluation of obtaining an appropriate regressor relationship. The detection data of the reflected wavelengths 104, 204, 205 obtained under laboratory conditions by the detection unit can be at least partially analyzed using machine learning. This can make it possible to obtain highly reliable information regarding the regressor relationship indicating the biomarker concentration in a specific physiological constitution of a specific body part (such as a finger).

[0087] The data unit 105 can be implemented within the device 100. However, the data unit 105 can be realized by the input / output interface of the device 100, and the data can be received and / or transmitted between the device and a remote data unit that stores the data.

[0088] FIG. 3 shows a schematic diagram of a figure showing the detected signals of different wavelengths 104, 204, 205 under pressure fluctuation I and under substantially constant pressure II according to an exemplary embodiment of the present invention.

[0089] In addition to the above measurements under pressure fluctuations shown in FIG. 2, the processing unit 103 is further configured to determine at least one measured value of the signal strength SS of the reflected light waves 104, 204, 205 during a particularly constant arrangement of the detector unit 102 with respect to the body part 110, and the measured value defines at least one measurement regressor (Rm). The value of the signal strength SS can be between 0 and 224000 nA in the example shown in FIG. 3. In FIG. 3, the pressure fluctuations over time t of the wavelengths 104, 204, 205, and thus the signal strength fluctuations, are shown under pressure fluctuation measurement I and non-pressure fluctuation measurement II.

[0090] The regressor relationship (RR) is further determined based on at least one determined characteristic regressor (Rc, Rcf) and at least one measurement regressor (Rm). During the arrangement of the detector 102 with respect to the body part, i.e., under (substantially) constant pressure, by further measuring the reflected wavelengths 104, 204, 205, the characteristic values of the reflected signals of the specific wavelengths 104, 204, 205 obtained under substantially constant pressure can define a measurement regressor Rm that can be used for the normalization of data regarding the current physiological constitution of the body part 110 and the calibration of the light source 101, for example, an LED. Since the measurement regressor Rm is further considered in the regressor relationship RR, an improved criterion for the regressor relationship RR with respect to the nominal regressor relationship indicating the biomarker concentration can be obtained.

[0091] In summary, during the measurement of the reflected light of the body part 110 under a given pressure fluctuation in the body part, each reflected wavelength 104, 204, 205 (e.g., red light, infrared light, blue light, green light, etc.) has a specific signal profile 201, 206 under a specific pressure fluctuation applied to the body part. It has been found that each signal profile 201, 206 under pressure fluctuation has respective specific signal intervals 202, 203, 207 in the signal profiles 201, 206 of the first reflected light wave. Furthermore, the specific regressor relationship RR can correlate much better with the biomarker concentration (e.g., glucose value) in the blood, taking into account the measurement regressor Rm obtained under substantially constant pressure shown in section II of FIG. 3.

[0092] As a mathematical relationship between the characteristic regressor Rc and, for example, the measurement regressor Rm, the following specific regressor relationships can be obtained: Rm1 / Rc1, Rm2 / Rc1, Rm1 / Rc2, Rm1 / ln(Rc1), ln(Rm1) / e Rc1 etc. The regressor relationship RR is correlated with the biomarker concentration in the blood, and the determined value of the regressor relationship is adapted to indicate the value of the biomarker concentration.

[0093] The measurement of the biomarker concentration using the device 100 can be performed as follows.

[0094] When the device 100 is covered by the body part 110 (for example, the fingertip, preferably the index finger or the ring finger), an optional fixing element of the device (rubber ring, rubber string or any other elastomer, rope, fastener, etc.) may be used to fix the body part 110 to the device for more accurate measurement.

[0095] Next, the measurement of the individual is started. Each person has different skin types and other physiological characteristics that can be evaluated by the device 100. According to the invention, the skin surface is pressed against the device with, for example, three consecutive pressures, whereby blood is squeezed out of the body part 110 (for example, the tip of the finger) and the body part shrinks slightly. For example, first, the pressure is gradually increased until, for example, the (for example, diode signal) is no longer distinguishable, which lasts for example about 10 seconds, then the pressure is gradually released for example for 5 seconds (for example, refer to the signal curve below section I in FIG. 3), and then the pressure occurs in a sequence where the entire process can be repeated, for example, two or more times. Then, the body part can remain on the device for example for 20 seconds under a substantially constant pressure (for example, refer to the signal curve below section II in FIG. 3).

[0096] Instructions to the person can be obtained from the display 106 of the device 100 (see FIG. 1).

[0097] Next, first, the validity and quality of the signal can be checked. Next, the physiological function of the finger skin can be examined based on the relationship between the regressor and the regressor relationship RR described above. The physiological function of the body part 110 is examined in each regressor relationship RR. Based on the measurement data, it is possible to determine the actual skin and subcutaneous characteristics based on the first part of the measurement I under pressure fluctuations, and to perform physiological normalization (FN) for the second part of the measurement II under a substantially constant pressure. Physiological normalization is used to normalize the data of the second part of the measurement by converting the values into a neural (universal) model (database), where all the obtained values have, for example, the same scale (unit). Based on the data of the regressor relationship RR, it is possible to determine the position of the actually measured regressor relationship RR within the multidimensional space of the database of the nominal regressor relationship that correlates with the concentration of the biomarker, for example, the blood glucose level.

[0098] The position of the measured regressor relationship within the multidimensional space of the database is determined, for example, based on clustering that determines the position of the statistical model within the spectral space of the model based on the data of the first part of the measurement. A more detailed classification of the measured regressor relationship can be provided by checking the relationship between signals of different wavelengths within a given measurement range.

[0099] Note that the term "comprising" does not exclude other elements or steps, and "a" or "an" does not exclude a plurality. Also, elements described in connection with different embodiments may be combined. Note also that reference signs in the claims should not be construed as limiting the claims.

[0100] [Description of Signs] 100 Device 101 Light Source 102 Detector Unit 103 Processing Unit 104 First Light Wave 105 Data Unit 106 Display 110 Body part 201 First signal profile 202 First specific signal interval 203 Further first specific signal interval 204 Second light wave 205 Third light wave 206 Second signal profile 207 Second specific signal interval I Pressure fluctuation measurement II Non-pressure fluctuation measurement SS Signal strength t Time Rc Characteristic regressor Rcf, Rcf2 Further characteristic regressors Rm Measurement regressor RR Regressor relationship

Claims

1. A device for determining the concentration of a biomarker in the blood of a body part in consideration of the physiological constitution of the body part, the device comprising: a light source for irradiating the body part with a first light wave; a detector unit for measuring the first light wave reflected from the body part; a processing unit coupled to the detector unit for receiving the measured first light wave; wherein: the processing unit is configured to: determine at least one characteristic value in a first specific signal interval in a first signal profile of the reflected first light wave that occurs during a predetermined pressure variation applied to the body part by the detector unit; the detector unit is capable of being pressed against the body part, the predetermined pressure variation being an increase or decrease in pressure between an initial pressure and a final pressure in the first signal profile, and the at least one characteristic value having a signal intensity (SS) of the reflected first light wave; the at least one characteristic value in the first specific signal interval of the reflected first light wave represents the physiological constitution of the body part and enables determination of the concentration of the biomarker in the blood.

2. The device according to claim 1, wherein the characteristic value further has a value of the slope of the time variation of the first signal profile in the first specific signal interval that occurs during the predetermined pressure variation applied to the body part by the detector unit.

3. The device according to claim 1 or 2, wherein the first light wave is selected from one of the group consisting of infrared light, red light, green light, and blue light.

4. The device according to any one of claims 1 to 3, wherein the first specific signal interval is defined by a characteristic slope, a flat region of a signal function, a sharp turn of the signal function, an inflection point, a minimum value or a local minimum, and a maximum value or a local maximum.

5. The processing unit is configured to: for each executed pressure variation, determine based on a plurality of repeated predetermined pressure variations occurring in the first specific signal interval of the first signal profile of the reflected first light wave; determine each characteristic value of the first specific signal interval in each predetermined pressure variation; determine an average characteristic value of the first specific signal interval determined in the predetermined pressure variation. The device according to any one of claims 1 to 4.

6. The at least one determined characteristic value defines at least one respective characteristic regressor (Rc), and the processing unit is configured to: determine a regressor relationship based on the at least one determined characteristic regressor (Rc), wherein the regressor relationship is correlated with the biomarker concentration in the blood, and the determined value of the regressor relationship indicates the value of the biomarker concentration, the device according to any one of claims 1 to 5.

7. The device further comprises a data unit having a dataset of a predefined regressor relationship correlated with each biomarker concentration, wherein the data unit is further configured to: compare the determined regressor relationship with a predefined regressor relationship, and when the determined regressor relationship is in the vicinity of the predefined regressor relationship, the biomarker concentration is derivable, the device according to claim 6.

8. The processing unit is further configured to: determine at least one further characteristic value having a further signal intensity (SS) of the reflected first light wave when a further first specific signal interval occurs in the first signal profile of the reflected first light wave during the predefined pressure variation applied to the body part by the detector unit, wherein the at least one further characteristic value in the further first specific signal interval of the reflected first light wave represents the physiological constitution of the body part, and the biomarker concentration in the blood is determinable, the at least one determined further characteristic value defines at least one respective further characteristic regressor (Rcf), and the regressor relationship is further determined based on the at least one determined further characteristic regressor (Rcf), the device according to claim 6 or 7.

9. The processing unit is further configured to: determine at least one measured value of the signal intensity (SS) of the reflected first light wave during a particular, in particular constant, arrangement of the detector unit with respect to the body part, wherein the measured value defines at least one measured regressor (Rm), ​ ​ ​ ​ ​ The regressor relationship is further determined based on the at least one defined characteristic regressor (Rc) and the at least one measured regressor (Rm), the device according to any one of claims 6 to 8.

10. The light source is configured to irradiate the body part with a second light wave, The detector unit is configured to measure the reflected second light wave reflected from the body part, The detector unit is configured to receive the reflected second light wave, The processing unit, When a second specific signal section in the second signal profile of the reflected second light wave during the predetermined pressure fluctuation applied to the body part by the detector unit occurs, determining at least one further characteristic value having the signal intensity (SS) of the reflected second light wave is configured to, The at least one further characteristic value in the second specific signal section of the reflected second light wave represents the physiological constitution of the body part, The at least one determined further characteristic value defines at least one respective further characteristic regressor (Rc2), The regressor relationship is further determined based on the at least one determined further characteristic regressor (Rc2), the device according to any one of claims 6 to 9.

11. The device is a portable device, the device according to any one of claims 1 to 10.

12. The biomarker is glucose, C-reactive protein (CRP), hemoglobin (HBC), cholesterol, LDL, HDL, fibrinogen, and / or bilirubin, the device according to any one of claims 1 to 10.

13. A method for determining the concentration of a biomarker in the blood of a body part in consideration of the physiological constitution of the body part, the method comprising: irradiating the body part with a first light wave, measuring the reflected first light wave reflected from the body part using a detector unit, determining at least one characteristic value in a first specific signal section in the first signal profile of the reflected first light wave that occurs during a predetermined pressure fluctuation applied to the body part by the detector unit comprising, the detector unit can be pressed against the body part, the predetermined pressure variation is an increase or decrease in pressure between an initial pressure and a final pressure in the first signal profile, and the at least one characteristic value has a signal intensity (SS) of the reflected first light wave, The method, wherein the at least one characteristic value in the first specific signal section of the reflected first light wave represents a physiological constitution of the body part and a biomarker concentration in the blood can be determined.

Images