DE102004039291A1 - Potential sensor for measuring electrical potential in skin surface, has generator determining division factor of voltage divider by derived clock pulse and allocating displacing component of artifacts to zero with output signal - Google Patents

Abstract

The sensor has a measuring electrode connected with an electrometer amplifier and an actuator. The actuator changes a distance between the electrode and a skin surface. The change is controlled by a generator (15). The generator determines a division factor of an input voltage divider by a derived clock pulse and allocates a displacing component of artifacts to zero with an output signal in an evaluation unit (16).

Description

The The invention relates to a potential sensor for measuring electrical Potentials on the skin surface with the help of sensor electrodes and a charge converter, which the Loads in adequate Tensions converted.

wobei die Ladung Q in eine Spannung U konvertiert wird, so dass die zu erwartende Messspannung u_a(t) mit der eingefügten Spannungsverstärkung ν_u der Primärelektronik sich ergibt als ua(t) = f(q•, CProbe, Ce, νu). (2) Potential measurements on surfaces are an early diagnosis in modern medical technology. For these measurements, the measuring electrodes should be positioned a short distance from the surface of the skin; an immediate attachment with an electrically non-conductive gel on the skin surface is also conceivable. The most important sensor parameter is the skin-electrode capacity, which lies in the very small pF range. As an evaluation system for the measurement technical recording of the charge changes at the measuring point of the subject an electrometer amplifier is preferably used, which is designed as a particularly sensitive voltage amplifier in Höchstohmtechnologie. The basic operation of a potential sensor and an open-type electrometer amplifier is particularly easy to analyze when the subject is selected as the ground (reference point). By this definition, the electrometer amplifier in the replacement circuit diagram can be represented by a series-connected capacitive voltage divider. The one series capacitor is formed from the body surface and the measuring electrode (C _sample ) and the other from the line capacitance C _L and the actual amplifier input capacitance C _V , which are collectively defined as the input capacitance C _e . Due to the technical structure, there is a resistor in parallel across the two series capacities, the size of which is in the high giga-ohm range. The entire system can therefore be represented as two series-connected RC combinations with different time constants and a subsequent voltage amplifier. The charge conversion happens after the relationship

wherein the charge Q is converted into a voltage U, so that the expected measurement voltage u _a (t) with the inserted voltage gain ν _{u of} the primary electronics results as u a (t) = f (q • , C sample , C e , ν u ). (2)

wobei die Gesamtkapazität C_ges. aus der Reihenschaltung von C_Probe und C_e gebildet wird. Das Vorzeichen gilt der Phasenlage der Ausgangsmessgröße und ist für die medizinische Auswertung irrelevant. In Verbindung mit dem Ladungsübertragungsfaktor S_q des Sensors und der Spannungsverstärkung ν_u = 1 wird die Ausgangsspannung am Elektrometerverstärker zu

The electrometer amplifier converts a charge according to the relationship due to its physical operation

the total capacity C _tot. is formed from the series connection of C _sample and C _e . The sign applies to the phase position of the output variable and is irrelevant for medical evaluation. In conjunction with the charge transfer factor S _{q of} the sensor and the voltage gain ν _u = 1, the output voltage at the electrometer amplifier is too

Since the charge transfer factor S _q is dependent not only on the charge to be measured q _sample but also because of (1) the sensor capacitance C _sample and its changes ± ΔC _sample , the entire output measurement signal becomes too high

bestimmt, wobei d der Abstand Messelektrode zur Hautoberfläche darstellt, und damit eine Funktion des Messabstandes wird, der beim Probanden nicht fixiert werden kann. Auf Grund der Bewegung Hautoberfläche zur Elektrodenoberfläche enthält das Messsignal neben der eigentlichen elektrischen Messgröße u(q) wegen Gl. (6) und Gl. (1) auch einen systematischen Fehler F(d ± Δd) im Ausgangssignal u_a(q) – die so genannten Artefakte. Diese Bewegungen sind stochastischer Natur und bei einem Messzyklus am Patienten kann aus Zeitgründen keine ausreichende Vorgeschichte des Fehlersignals erstellt werden kann, sodass bekannte statistische Messfehlerkorrekturverfahren nur unzureichend anwendbar sind. Für die zu messende Ausgangsspannung u_a(t) für einen Elektrometerverstärker, der Tiefpassverhalten aufweist, gilt demnach

mit τ_ges. als der in Reihe geschalteten Eingangs-RC-Kombinationen mit unterschiedlichen Übertragungszeitkonstanten. Zum Stand der Technik für die Realisierung können Arbeiten der Physical Electronics Group von der University of Sussex genannt werden (z. B.: „An ultra-low-noise electrical-potential probe for human-body scanning" in School of Engineering, University of Sussex, Brighton, Sussex BN1 9QT, UK und die Schrift „Remote detection of human electroencephalograms using ultrahigh input impedance electric potential sensors" in Applied Physics Letters, Volume 81, Nr. 1721, Oct. 2002, einer theoretischen Arbeit zum grundsätzlichen Messprinzip mit veränderbaren Kondensatoren von Lord Kelvin (1861) sowie die Patentschrift DE 006918190 T2 (Anmeldedatum 9. Mai 1995).It can be seen from the equation (2) that the measurement output voltage depends both on the charge / charge change quantity actually of interest on the skin surface and on the capacitance changes ± ΔC _{sample of} the measurement object. The capacity C is determined by the design equation

where d represents the distance measuring electrode to the skin surface, and thus becomes a function of the measuring distance, which can not be fixed in the subject. Due to the movement skin surface to the electrode surface contains the measurement signal in addition to the actual electrical quantity u (q) because of Eq. (6) and Eq. (1) also a systematic error F (d ± Δd) in the output signal u _a (q) - the so-called artifacts. These movements are stochastic in nature, and a measurement cycle on the patient can not provide a sufficient history of the error signal due to time constraints, so that known statistical measurement error correction methods are insufficiently applicable. Accordingly, the following applies to the output voltage u _a (t) to be measured for an electrometer amplifier which has low-pass behavior

with τ _sat . as the series-connected input RC combinations with different transmission time constants. The state of the art for the realization of this work may be mentioned by the University of Sussex Physical Electronics Group (eg: "An ultra-low-noise electrical-potential probe for human body scanning" at School of Engineering, University of Sussex, Brighton, Sussex BN1 9QT, UK and the paper "Remote detection of human electroencephalograms using ultrahigh input impedance electric potential sensors" in Applied Physics Letters, Volume 81, No. 1721, Oct. 2002, a theoretical work on the principle principle of variable Capacitors of Lord Kelvin (1861) and the patent DE 006918190 T2 (Filing date 9 May 1995).

Possibilities for solving such metrological task are circuit measures at the electrometer amplifier input such. B. phase-sensitive evaluation systems or electronic chopper amplifiers. Because of the extreme high resistance of the electrometer input (R _e ≥ 10 ¹² ohms) and the very low input capacitances (C _sample and C _e ≈ 0.5 ... 5.0 pF), such circuit measures can only be implemented for selected applications. Downstream chopper systems or passive filter networks also did not achieve the desired success in extremely high-impedance primary measurement systems because of their high own time constants (τ ≈ ≥ 0.5 ... 1.5 s). A signal override of the primary electronics would distort the output signal so that the actual measurement signal waveform can not be reconstructed. Determining the capacitances of the series-connected input voltage divider C _Probe and C _e for an amplitude evaluation by an alternating measurement voltage fed in on the input side in order to determine the division factor is only possible academically because of the very low capacitance values and the extreme high resistance. The same applies to the measurement of the current sensor capacitance of C _Probe by means of an input alternating voltage.

It So there is the task, the temporal distance and motion-dependent distortions (Artifacts) of the original Compensate signal.

According to the invention this Problem solved by the features of claim 1, the dependent claims advantageous embodiments to the invention.

The Invention will be explained below with reference to a single drawing. there in detail mean:

1

distance change Skin surface / measuring electrode

2

distance Skin surface / measuring electrode

3

charge potential (Measured variable)

4

Measuring surface, (e.g. Scalp)

5

measuring electrode

6

Electrometer amplifier

7

primary electronics

8th

Divider clock signal

9

Primary output signal

10

Actuator movement

11

Potential reference point

12

Actuator control signal

13

actuator

14

synchronous clock

15

generator

16

evaluation

17

output

erstellen, mit dem gezeigt werden kann, dass durch Artefakte bedingte Abstandsänderungen 2 im Sensor ausgeblendet werden können. Der Aktuator 13 wird durch das Steuersignal 12 angeregt, welches von einem Muttergenerator 15 geliefert wird, der gleichzeitig mit einem Synchronsignal 8 die Primärelektronik 7 steuert und einen abgeleiteten Synchrontakt 14 an die Auswerteeinheit 16 liefert, da es für den Berechnungsvorgang erforderlich ist, den Zeitpunkt der ermittelten Messspannungen [U₁(d₁) und U₂(d₂)] 9 in Bezug auf die Stellung der Messelektrode 5 zum Messobjekt 4 exakt zu kennen. Zwei a priori Voraussetzungen für das Berechnungsprozedere sind unbedingt erforderlich:

• 1. Die Größen U₁(d₁), U₂(d₂) und Δd müssen ungleich null sein;
• 2. Die Signalspannung U_Sign. muss während des Messvorganges U₁(d₁) und U₂(d₂) unverändert bleiben;

The basic idea of the invention is based on the connection that a charge Q at a capacitance C is converted into a voltage U [s. Eq. (1)]. This capacitance C is not constant because of the artifacts and by applying a known quantity Δd, the fluctuations (artifacts) that are present in the measuring voltage can be eliminated by higher-order linear systems of equations. The starting point for these operations is equation (6). The electrode distance 2 to the skin surface 4 is through an actuator 13 in the range Δd 10 changed by the measuring electrode 5 , preferably mechanically coupled to the primary electronics 7 , in a defined way relative to the skin surface 4 is vibrated back and forth (movement 10 , if possible in the mm range). This is preferably done with a frequency that is considerably higher than the maximum signal frequency to be measured. An advantageous method, because a low-electrical implementation is possible, z. B. a pneumatic actuator. By the movement changes over the distance 2 the voltage 3 in the rhythm of the actuator movement 10 , Here d _{0 is} the base distance skin surface 4 and measuring electrode 5 be; d _{1 represents} the reduced and d ₂ the increased distance. With the simplification ε ₀ · ε _r · A = k and the relationships d ₂ = 2d ₀ - d ₁ and d ₂ = d ₁ + Δd can be calculated from the measured voltages U ₁ (d ₁ ) and U ₂ (d ₂ ) and the known quantity Δd the system of equations

which can be used to show that artifact-related distance changes 2 can be hidden in the sensor. The actuator 13 is by the control signal 12 stimulated, which of a mother generator 15 delivered simultaneously with a synchronizing signal 8th the primary electronics 7 controls and a derived synchronous clock 14 to the evaluation unit 16 Since it is necessary for the calculation process, the time of the determined measuring voltages [U ₁ (d ₁ ) and U ₂ (d ₂ )] 9 with respect to the position of the measuring electrode 5 to the measurement object 4 to know exactly. Two a priori requirements for the calculation procedure are absolutely necessary:

• 1. The quantities U ₁ (d ₁ ), U ₂ (d ₂ ) and Δd must not be equal to zero;
• 2. The signal voltage U _Sign. Must remain unchanged during the measuring process U ₁ (d ₁ ) and U ₂ (d ₂ );

requirements so far fulfilled are that they do not make any significant error contribution to the measurement deliver.

entsteht eine nichtlineare Eingangskennlinie mit hyperbolisch ähnlichem Verlauf, die einen multiplikativen Fehler in die Ausblendung der Artefaktebewegung bringt. Ihn zu eliminieren gelingt nur, wenn die Kapazi tätswerte C_Proobe und C_e bekannt sind oder ihr Teilungsfaktor. Ersteres ist wegen o. g. Gründe nicht sinnvoll. Der Teilungsfaktor kann bestimmt werden, indem aus dem Teilertaktsignal 8 in der Primärelektronik 7 ein Synchronsignal erstellt wird, das als getaktetes Referenzsignal dem Eingang des Elektrometerverstärkers 6 aufgeschaltet wird und mit dem Reziprokwert des gemessenen Signals 9 in der Auswerteeinheit 16 verrechnet wird. Die Ermittlung der wirklich existierenden Abstände 2 (d₁ und d₂) ist als Berechnungs-Nebenprodukt möglich und denkbar, aber nicht notwendig. Systemtheoretisch handelt es sich bei der Bestimmung des Teilungsfaktors um die Berechnung des Amplitudenfrequenzgang zweier in Serie geschalteter Tiefpässe mit unterschiedlichen Zeitkonstanten [s. Gl. (3)] in drei Fixpunkten. Die Elektrodenflächen 5 ist mitbestimmend für die Größe der Sensorkapazität [s. Gl. (6), in der A die Fläche der Messelektrode 5 bildet], die aber nach einmaliger Größenfestlegung des Potenzialsensors konstant bleibt. Die Berechnung des Ausgangssignals u_{a Mess.} 17 wird vorzugsweise mittels eines Mikroprozessors und eines geeigneten Algorithmus durchgeführt.The most important problem for the masking of artifact errors is the input capacitance C _{e of} the electrometer amplifier 6 , This is of the order of magnitude of the sensor capacitance, amounts to only a few picofarads, and forms with it a capacitive voltage divider. Due to the determination of the replacement capacity

The result is a nonlinear input characteristic with a hyperbolic similar course, which brings a multiplicative error in the suppression of artifact movement. To eliminate it succeeds only if the capacitance values C _Proobe and C _{e are} known or their division _factor . The former is not useful because of the above reasons. The division factor can be determined by dividing from the divisional clock signal 8th in primary electronics 7 a synchronizing signal is generated, which as a clocked reference signal to the input of the electrometer amplifier 6 is switched on and with the reciprocal of the measured signal 9 in the evaluation unit 16 is charged. The determination of the really existing distances 2 (d ₁ and d ₂ ) is possible as a calculation by-product and conceivable, but not necessary. In terms of system theory, the determination of the division factor involves the calculation of the amplitude frequency response of two series-connected low-pass filters with different time constants [s. Eq. (3)] in three fixed points. The electrode surfaces 5 determines the size of the sensor capacity [s. Eq. (6), where A is the area of the measuring electrode 5 forms], which remains constant after one-time size determination of the potential sensor. The calculation of the output signal u _{a Mess.} 17 is preferably with performed by a microprocessor and a suitable algorithm.

Claims (5)

Potential sensor for charge measurement at the skin surface ( 4 ) with a measuring electrode ( 5 ) with an electrometer amplifier ( 6 ), characterized in that the measuring electrode ( 5 ) on an actuator ( 13 ), the actuator ( 13 ) the distance ( 2 ) between the measuring electrode ( 5 ) and the skin surface ( 4 ), the distance change ( 10 ) from a generator ( 15 ) is controlled by the generator a clock signal ( 8th ), the primary electronics ( 7 ) determines a division factor for the capacitive input voltage divider at defined times and an evaluation unit ( 16 ) the division factor with the useful signal ( 9 ) resolves the artifacts to zero. Potential sensor according to claim 1, characterized that the actuator is preferably designed as a pneumatic oscillator. Potential sensor according to claim 1 and 2, characterized in that the measuring electrode ( 5 ) and the electrometer amplifier ( 6 ) form a unit. Potential sensor according to claim 1 to 3, characterized in that the electrometer amplifier ( 6 ) represents in its transmission range a low-pass system only 2nd order. Potential sensor according to claim 1 to 4, characterized in that the output signal ( 9 ) at defined times with the actuator movement ( 10 ) is synchronized.