Background and Summary of the Invention
The invention refers to implantable pacemakers as those described in European application No. EP-A2-0 089 014 (and corresponding copending U.S. application Ser. No. 06/474,241 in the name of the same applicants) wherein the stimulation/inhibition rate of the pacemaker depends, by a programmable algorithm, on the respiratory rate of the patient, detected through the measurement of the impedances of a subcutaneous part of the patient's chest. Said European application No. EP-A2-0 089 014, and its U.S. counterpart, concerns a unit that extracts a clean signal, e.g. a squared one, from the signal coming from the body impedance variation detecting unit. Said clean signal shows a transition every time that the input signal exceeds a fixed and preferably programmable value. In the structural arrangement of the unit, the suggested solution allows to detect the amplitude variations of the input signal only if said variations are positive in comparison with the base line. The same unit is then able to detect only slow variations of the base impedance. Hospital practice has shown that in some cases the variations of the input signal amplitude of said unit can be both positive and negative with regard to the base line. The present invention relates to a new signal extracting unit for pacemakers, of the type above mentioned, capable to meet said needs and to produce a clean output signal, e.g. a squared one, that shows a transition every time that the difference between a minimum value and a maximum value or a maximum value and a minimum value of the input signal exceeds a fixed and preferably programmable value. Moreover the unit according to the invention is able to perform both slow and fast and recurrent variations of the base line. In consideration of the fact that such a signal extracting unit has proved to be very useful and reliable, the invention proposes further an alternative structural arrangement of the same unit to make it suitable to be assembled in physiologic pacemakers as those described in the Italian patent application No. 12421 A/84 of Feb. 7, 1984 by the same applicants, and corresponding copending U.S. application Ser. No. 06/659,542, wherein the stimulation/inhibition rate depends, by a special programmable algorithm, on the patient's pulmonary minute ventilation (1/min).
Brief Description of the Drawings
The features of the circuit of said unit and its advantage are apparent in the following description of some preferred embodiments of same, shown by way of non-limiting example in the attached drawings, in which:
FIG. 1 shows the circuit of the unit processing the output voltage variations of the analyzer and convertor circuit which calculates the impedance variations of a part of the patient's body with a sample rate such as to allow to detect the respiration as described in the mentioned U.S. application Ser. No. 06/474,241;
FIGS. 2 and 3 show the features of the circuit signals which are present in the circuit according to FIG. 1;
FIG. 4 shows the circuit of FIG. 1 modified in order to use the physiologic pacemaker according to Italian patent application No. 12421 A/84 depending on the patient's pulmonary ventilation;
FIG. 5 shows the features of some of the signals present in the circuit in FIG. 4.
Description of the Preferred Embodiment
With reference firstly to FIG. 1, 10 is the reference numeral for the unit already described in application Ser. No. 06/474,241 which detects the impedance variations ZL of the patient's body by producing sample pulses I.sub.Test lasting 122 .mu.s and with a repetition period To=125 ms (timing A in FIG. 2). The peak value of the voltage V.sub.ZL =I.sub.Test .multidot.ZL is stored for 366 .mu.s by a fast "Peak and Hold" (P/H). Said P/H can be avoided if the differential gain (U1) at the unit input according to the present invention (as described hereafter) is sufficiently fast.
In the present scheme in order to save energy, a slow differential amplifier is used in combination with a fast and low consumption P/H. The stay time of the peak voltage (366 .mu.s) at the P/H output is controlled by the timing Q of FIG. 2 with the same rate as signal A.
V/H is the P/H output voltage controlled, as said, by Q shown in FIG. 2. The voltage V.sub.H is applied to the noninverting input of amplifier U1 and when Q is high, via the differential amplifier U1 with gain A.sub.O and switch SW1, closed when Q is high, it determines the charge of condenser Cy. Differential amplifier U1 is fully supplied with operating voltage only when Q is high in order to save energy. Q has the same rate and timing of the sampling. Let us take a general sample cycle called "m". At the end of "m" time interval of 366 .mu.s, during which Q remained high, Cy is charged with U1 output voltage. The inverting input of amplifier U1 receives the charge voltage of a condenser Cy. ##EQU1## where -V/2 is a voltage fixed for the purposes given as a reference and .DELTA.V.sub.Cym the difference, positive or negative, from said reference. It has to be taken into account that all increases .DELTA. can be either positive or negative.
When Q goes back to low, switch SW1 is opened, therefore Cy is disconnected and charged with V.sub.Cym.
Let's now consider the zeroing logic group LU. There is no detailed description of said group as it can be easily realized by any expert in the field according to the description of the functions given hereafter, taking into consideration the diagram of the timings of FIGS. 2 and 3. LU has two inputs: COMP indicates one of them wherein a comparator U2 compares V.sub.Cym and -V/2; GO indicates the other one wherein timing B has the same rate of Q but slightly delayed. LU has two outputs: one is called EN and when it is high it closes switches SW2, SW3, SW4, the other, CH/DIS commutes switches SW5, SW6, SW7 to the left when it is low and to the right when it is high. EN output goes high with B rising edge and it stays there until there is a change in the comparator output. CH/DIS output is low when COMP input is high and high when COMP input is low. After a Cy charge increase and the opening of SW1 for the continuous action of comparator U2, switches SW5, SW6, SW7 are positioned to the right or to the left according to COMP input level that is according to .DELTA.V.sub.Cy sign with reference to -V/2. With the arrival of B timing pulse, its rising front triggers off LU which, in its turn, closes SW2, SW3, SW4 allowing as a first consequence, the voltage pulses produced by I.sub.O+ or I.sub.O-,lasting .tau.O=3 .mu.s, with I.sub.O amplitude and timed by CK.sub.o (FIG. 2) to charge or discharge condenser Cy to cancel .DELTA.V.sub.Cym. The canceling takes place after N.sub.m pulses being ##EQU2## and, at this point, the discharge is blocked as the sign of Cy voltage changes with regard to -V/2 reference voltage on comparator U2, the output of which operates on the logic group LU. During Cy charge adjustment, current pulses with the same amplitude, duration and recurrence rate as those which adjusted Cy, but produced by another I.sub.O+ and I.sub.O- source connected to Cx, modify in the opposite way the charge of Cx that during the "m" cycle increases its voltage ##EQU3##
At the end of "m" cycle and at the beginning of "m+1" cycle, Cx voltage becomes: ##EQU4##
According to (1), in "m+1" cycle the situation will be: ##EQU5##
In "m" cycle, V.sub.H voltage increases that is:
and by replacing in (5) the data given by (6) and (4) ##EQU6##
Going back to (1), simplifying and gathering the data we obtain: ##EQU7##
If the capacity and the gain values of the differential amplifier U1 are such that: ##EQU8##
When (8) is verified Cy voltage referred to -V/2 is equal to the gained difference between the present and the previous V.sub.ZL sample. If we consider (2), N.sub.m is proportional to said .DELTA.V.sub.Hm difference or, as T.sub.o is constant, to the incremental ratio (.DELTA.V.sub.Hm)/T.sub.o.
Going back to group LU, it is now clearer that, according to the output state of comparator U2 and, therefore, to V.sub.Cy value vis-a-vis -V/2, SW5 and SW6 are positioned so as to supply a current with the sign suitable to bring V.sub.Cy back to a value close to -V/2.
The rising edge of B triggers off V.sub.Cy adjustment action by adding or subtracting I.sub.O .tau..sub.O charge quantities in Cy as the same edge closes, through EN output, switches SW2, SW3, SW4 that will reopen when comparator U2 shows the sign change of the difference V.sub.cy -(-V/2)=.DELTA.V.sub.Cy.
The action of the first two switches is evident. SW4, according to the position of SW7 that depends, all things considered, on the sign of the incremental ratio, instructs N.sub.m pulses of CK.sub.o to increase (N.sub.UP) or decrease (N.sub.DW) a UP/DW programmable and saturating register of finite length indicated with U3 in FIG. 1. By UP/DW saturating register, we mean a register counting in the two directions which stops when it reaches one of the two count limits (N.sub.max or O) without accepting further pulses.
In order to make clear what said so far, and in particular function of register U3 and the circuits connected to it, we take an example shown in FIG. 3. We can see the behaviour of V.sub.ZL with its samples. Corresponding to it, just below, we see the behaviour of V.sub.Cy referred to -V/2. If we consider equation (9) and the following considerations we notice that the envelope of V.sub.Cy diagram represents the approximate derivative, at a distance of a sample interval of ZL.
Each V.sub.Cy sample with positive V.sub.Cy (or negative) is associated with a N.sub.UP (or N.sub.DW) which determines its size. These numbers are added algebraically in the register, starting from a saturation state until the next one. The result corresponding to each sample is N.sub.COUNT which, once it has reached a higher or lower saturation, it stays there until subsequent subtrahends or addends bring it back within the range with O and N.sub.max as limits (programmed from the outside as shown in FIG. 1). In the example of FIG. 3 N.sub.max is 32.
The indication of upper (OVERFLOW) or lower (UNDERFLOW) saturation by register U3 are given to a flip-flop U4 that, through the above described process, produces an output signal V.sub.OUT, depending on V.sub.ZL. Signal V.sub.OUT is high when N.sub.COUNT reaches O and low when N.sub.COUNT reaches N.sub.max value. FIG. 3 shows the relation between the behaviour of V.sub.ZL and that of V.sub.OUT pointing out the changes in state of the latter one. V.sub.OUT depends, as said, on the number of pulses necessary for Cy discharge. The conclusion we draw from FIG. 3 is the following. For each increase of V.sub.ZL greater than a given V.sub.TH value, starting from its maximum value, there is a negative transition of V.sub.OUT. V.sub.OUT transitions, as already said, take place at the same time of the sample, after exceeding V.sub.TH value. The threshold voltage is: ##EQU9##
Although ZL is a particular impedance, for the purposes of the present invention it can be defined as a threshold impedance, considering that V.sub.ZL peak=(ZL.I.sub.TEST) peak: ##EQU10##
This threshold can be regulated by N.sub.max which is programmable from outside.
Taking into account the above algorithm, the operation of the device can be thus summarized. There is a V.sub.OUT transition each time that the difference between a minimum (or maximum) impedance value and a maximum (or minimum) value of the following sample exceeds Z.sub.TH fixed value. Cases wherein the difference between a minimum (or maximum) value and a maximum (or minimum ) value of the subsequent samples are lower than Z.sub.TH value are entirely neglected. This statement solves the case of spurious signals (such as those due to the heart beat) overlapping the main signal. In the specific case of application to the transthoracic impedance measurement, the device exclusively detects the relatively large and repeated variations depending on respiration.
In order to be used in the physiologic pacemaker described in Italiam patent application No. 12421 A/84 mentioned in the introduction, wherein the stimulation/inhibition rate of the pacemaker is controlled by the patient's pulmonary minute ventilation (1/min), the circuit of FIG. 1 can be modified as shown, as an example, in FIG. 4, wherein components SW7 and U3-U4 have been eliminated and at every 125 ms sample interval, pulse trains come out of switch SW4, their quantity being proportional to the body impedance variations between consecutive samples. FIG. 5 shows the behaviour of V.sub.ZL with the relevant samples. Corresponding to it, just below, we can see the behaviour of V.sub.Cy referred to -V/2. Further below, the timing of V' (.DELTA. ZL) is indicated for the first five samples (N.sub.DW and N.sub.UP timing of FIG. 3). Signal V' (.DELTA. ZL) coming out of switch SW4 is sent to an electronic unit 100, for example a microprocessor or other type with low energy consumption and which can be miniaturized, unit which knowing the repetition period (125 ms) of the input pulse trains and the rate (122 .mu.s) of the same train pulses, produces a signal Vm (.DELTA. ZL) that is the average of the variation speed of signal V' (.DELTA. ZL) and is proportional to the patient's pulmonary minute ventilation (1/min).
Signal V.sub.OUT in the circuit of FIG. 1 and signal Vm (.DELTA. ZL) in the circuit of FIG. 4, is sent to a programmable correlator, not shown in the Figures, which produces a pacemaker stimulation/inhibition rate that is proportional to the features of the input signal and suitable for the physiologic demand of the patient.
It is understood that the circuit of FIGS. 1 and 4 can be used in implantable and/or portable devices other than the pacemakers, such as devices physiologically driving the operation of an artificial organ, for example an artificial heart, or in therapeutic and/or diagnostic devices. Lastly, it is understood that the description is referred to a preferred embodiment of the invention, which may be changed or modified, in its constructive realization, without departing form the inventive principle above described and hereafter claimed.