FR2473874A1 - METHOD AND APPARATUS FOR MONITORING AND STORING CARDIAC ACTIVITY SIGNALS WITH A DATA PROCESSOR - Google Patents

Abstract

THE INVENTION RELATES TO A PROCESS AND AN APPARATUS FOR MONITORING AND STORING SIGNALS OF HEART ACTIVITY, INCLUDING A DATA PROCESSOR. THE DEVICE INCLUDES A MEMORY 26 WITH SEVERAL MEMORIZATION ZONES 26A, 26B, 26C, 26D, A DEVICE 12 WHICH SAMPLES AT PERIODIC INTERVALS A SIGNAL OF CARDIAC ACTIVITY IN REAL TIME, AND WHICH STORES THE SAMPLES OF THE SAMPLES IN ITS POSITIONS MEMORY ZONE, AND A DATA PROCESSOR 14 WHICH PROCESSES THE SAMPLED AND STORED SIGNALS AND WHICH STORES THEM AFTER THEIR PROCESSING IN THE SECOND MEMORY ZONE FOR DISPLAY. THE INVENTION APPLIES IN PARTICULAR TO THE SURVEILLANCE OF HEART DISEASES NORMALLY GOING ON THEIR OCCUPATIONS.

Description

The present invention relates to a settlement apparatus

  and storing data indicative of cardiac activity.

than a sick person.

  The need for recordings of

  heart, and in particular electrocardiograms

  over a relatively long period of time in

  Seeking their occupation has been recognized in the prior art as described in the article entitled "The Use of R-R Interval and Different Histograms in Classifving Disorders of Sinus Rhythm" by P.M.M. Cashman published in the January 1977 issue of "Journal of Medical

  Engineering Technologv "The essential objective of these

  is to enable the identification of disturbances

  infrequent and transient heart rhythm

  which may be important for the diagnosis of

  who have vague or intermittent symptoms,

  for example, vertigo, syncope and fainting

  ments. A recording for an extended period

  offers the doctor the possibility of detecting short

  disrythmic events which occupy only a small percentage

  age of the total recording time. These disryth-

  are considered singularities of a basic rhythm. In general, the physician is interested not only in the basic rhythm but also in the particular dysrhythmic events which may consist of slow responses of the heart to influences such as pharmaceutical treatment or psychological stress for prolonged periods. In this respect, it can be

  desirable to compare the long records of

  different patients or the same patient at different

  your. The use of the cardiac control devices of a standing patient, for example the Holter Recorder

  is now well known. According to this technique, electricity

  trocardiogram of a patient is recorded for at least

  24 hours. The difficulty posed by the use of a

  Holter recorder is that it delivers a large volume

  information that requires processing and analysis.

  This analysis is usually long and expensive. The

  examination and treatment time may be reduced by

  and displaying the data at accelerated speeds, for example multiples of 25.32.60-etl2O of the normal reproduction frequency. To view the car-

  dies, and in particular the QRS complex with a clear

  If necessary, it may be necessary to use recording devices with frequency responses exceeding 12 KRz. This is possible with ultraviolet recorders, fiber optic recording oscilloscopes or jet recording devices

  of ink that are all very coiling.

  Various devices have already been suggested for processing and displaying cardiac activity data of a patient. A device of this type may include a detector that automatically detects a disrythmia and which, in response thereto, stores and displays a

  sample of the patient's electrocardiogram, directly

  on a paper or screen. In addition, an oscillo-

  gram can be provided by a memory oscilloscope or

  a Polaroid camera so that successive lengths

  of the patient's electrocardiogram are displayed

  one below the other. Each segment or

  waveform is proportional to the interval between beat and the right edge of the waveform gives a continuous record of the trend of the interval

  R-R. One of the most common methods of representing

  Electrocardiogram data is to obtain a tendency of the heart rate in which the interval

  R-R is measured continuously and the interval data is

  the R-R are segmented into continuous periods. In by-

  In particular, the average frequency or beat interval is set to obtain a mean frequency or mean interval display during a series.

  of contiguous periods. To resolve the indi-

  in a trend line, it may be preferable to plot the instantaneous frequency in the form of a

trend.

  In addition to the display methods mentioned above, the need exists to identify and indicate the isolated occurrence of an abnormal or ectopic beat. In general, ectopic beats are detected and are counted during the course of the cardiac control period. A

  This device was designed to detect premature beats

  tured. The degree of urgency of a beat can often be

  set by the operator to make a selection of

  to be detected. The output signals of these detections

  are applied to a counting circuit, a recorder

  trending or any form of alarm device depending on the desired application. Various techniques have been implemented to detect the abnormal form of the electrocardiogram signal of a patient, including

  integrators, filters and digital devices

  recognition of form. It should be noted that normal electrocardiogram signals are analyzed for

  determine the limits of appearance of the peaks of comple-

  QRS and to compare these limits with the incoming signal

  to determine whether the electrocardiogram signal

  falls within these normal and predetermined limits.

  In more elaborate devices, electrocardiogram-

  me normal of the patient is detected and memorized. In subsequent control, the normal patient signal is used as a reference for comparison with current cardiac activity data. U.S. Patent No. 4,115,864 discloses a cardiac control device of US Pat.

  this kind, using a calculator to control the memo

  of the patient's normal signal, to treat the patient's

  electrocardiogram signal entering into digital segments

  to compare these segments with segments

  correspondents of the normal and already memorized electrocardiogram of the patient. If the stored normal signal does not correspond to the current signal, an alarm or detection circuit is controlled and the number of appearances of this ectopic signal is stored in the computer memory. This device can give a past tendency of ectopic beats

  in the form of a histogram or a trend line.

  In any electrocardiogram data processing,

  it is necessary to use discriminatory techniques

  to extract a trigger pulse in each cardiac cycle. In most cases, the circuit

  detection is triggered on the R-wave and it is therefore necessary

  re to distinguish the R wave from the QRS complex from the rest of the signal.

  It should be noted that it may be necessary in the presence of a loud background noise, to interrupt the control during the duration of the high level background noise. In other cases, the detection circuit may give an alarm so that a

operator can intervene.

  With respect to the desire to perform long-term control of a patient's cardiac activity, continuous control by the Holter device results in complete registration for a relatively low price. On the other hand, data processing and compaction methods for simpler and more limited display often require a more complex, higher priced device. To avoid this disadvantage, it has been suggested that the data could be presented in the form of a RR interval histogram as shown in Figure 1. A histogram shows cardiac activity data taken during a period in a compact form. according to which the successive intervals between the waves R are calculated and are classified according to their duration. As shown in Figure 1, the X axis is graduated in seconds corresponding to the R-R interval while the Y axis gives the

  number of beats appearing in each interval.

  This presentation leads to a large reduction of the data, in a visual way, allowing an easy comparison between the histograms of the same patient at different times.

  or between different patients. The histogram of the intervals

  the R-R is formed by producing a group of columns or

  of boxes each corresponding to a range of values of

  between beats (or R-R). When each electrocardiogram complex is detected, the time interval that separates it from its predecessor is measured and the total is increased in the appropriate box. A current histogram can comprise a hundred boxes each representing a duration of twenty milliseconds, a total range of intervals R-R from 0 to 2 seconds. A box capacity of 4095 beats (a 12-bit word) is used to control a

  normal heart for about four hours.

  Figure 2 shows a variant of the Ithistogram of the R-R intervals in the form of a difference histogram.

  intervals R-R. The latter is formed in a similar manner

  except that the amount indicated along the X axis, as shown in Figure 2, is the value

  whose R-R interval varies between successive beats.

  As shown in Figure 2, the central box is designated

  zero, that is to say a zero difference between

  successive steps and a hundred columns show a range between 1second to 1second with a box width of 20milliseconds. The average of the different intervals

  is always very close to zero. This histogram of dif-

  gives an indication of how heartbeat varies, the interval histogram being

  the measurement of the dispersion of the frequency of beats

  turn of the average value. It appears that the use of these two types of histograms is for the physician

  a powerful tool for diagnosing the patient's heart.

  U.S. Patent No. 4,146,029 discloses

  a device implanted in the patient's body to dissect

  think a medication. The device is made with a

  microprocessor which controls the operations, during

  which each QRS complex of the patient's heart is detected as to its length, as well as the interval that separates the

  complex. A comparison program of the length of the com-

  QRS plexus with an acceptable length is planned to de-

  terminate the validity of each QRS complex and additionally, to first calculate the interval between complexes

  successively and compare the measured value with a

  nude or standard for a particular patient. More particularly

  The program compares the duration to determine

  how much the measurement intervals are shorter than the

  Normal time and, depending on the decrease in duration, that is to say the increase in heart rate, the microprocessor controls the dispensing device to change the dosage given to the patient. The apparatus also contains a program for measuring the R-R interval and for accumulating the average of the R-R intervals for a given period, for example one hour. The averages of the intervals are compared with known standards for ordering

  variable way the delivery of the medication to the patient.

  An object of the invention is therefore to propose a new

  apparatus for monitoring and memorizing activity signals

  using a data processor, for example

  ple in the form of a digital microprocessor.

  Another object of the invention is to control the process

  the data in a more efficient manner for evaluating and storing signals representing cardiac signals of the patient. Another more particular object of the invention is to

  to propose a device for controlling and memorizing

  indicating a cardiac activity of a patient, the

  made in the form of a treatment apparatus

  of data that works in a first time frame-

  the actual relatively slow to sample signals from the patient's heart and in a second time frame of calculator, relatively fast to process and evaluate

the sampled signals.

  According to these and other objects, the invention relates to

  therefore an apparatus for monitoring and storing electrical signals indicative of a cardiac activity of a patient, comprising a memory with first and second groups of storage positions, a circuit operating in a real time frame for sampling at a regular frequency cardiac activity of the patient and to memorize

  these sample signals in corresponding positions

  from the first group of memory positions, and a process

  a data processor which processes the signals stored in the first group and stores the processed signals in determined positions of the second group. The memorization of the signals in the second group of positions is performed

  in a second time frame of the calculator, at a frequency

  higher than the real-time sampling rate.

  One embodiment of the invention comprises a hor-

  sampling box which delivers a sampling clock signal to a sampling circuit in order to

  take samples of the cardiac activity of the patient.

  A second computer clock circuit delivers signals

  the computer clock to at least a part of the program

  data processor so that the processing of the activity

  heart rate is performed in a time frame of the

relatively fast.

  According to another aspect of the invention, the processor of

  data includes a first part or a first

  tif that operates in response to the sample clock signal.

  by evaluating the cardiac activity signals to determine the existence of a valid beat. In one embodiment, the cardiac activity signals

  are assessed to determine segments and to

  then a series of these segments to determine whether

  a valid heartbeat was controlled.

  In yet another aspect, the data processor

  has another part running in the temporary frame

  calculator to treat the detected heartbeat and, in particular, to determine the interval

  between that beat and the next beat. Representatives

  between the intervals between successive beats and the frequency of the beats are stored in another position of the memory. These stored data are

  evaluated; to determine whether the heartbeat is

  appeared in a regular rhythm or else it is an ano-

  malie. If this abnormality is detected or if the patient has

  by a manual switch given an indication of an anomaly

  dregs or cardiac discomfort, the data processor

  transfers a series of sample heart samples

  born in yet another position of memorization so that

  it is stored there and then read and displayed.

  Other features and advantages of the invention

  will appear in the following description.

  In the accompanying drawings, given solely by way of non-limiting example

  Figures 1 and 2 show histograms of

  R-R intervals and different intervals during a period

  3A and 3B are respectively a simplified, high level functional diagram of the basic elements of the cardiac control device according to the invention, and a perspective view of the portable data entry unit incorporated in the device. of Figure 3A,

  FIGS. 4A, 4B, 4C, 4D and 4E are detailed diagrams

  parts of the device represented in a general way.

  In FIG. 3A, FIG. 5 shows the direct access memory areas constituting a memory of the portable data entry device of FIG. 3A.

  Figure 6 is a high-level organization chart of the

  deferred time program stored in the permanent memory

  executed by the central processing unit of the

Figure 3A.

  Figure 7 shows the different subprograms of

  service and the way in which they are associated to coordinate

  In the real-time tasks of the device of FIG. 3A, FIGS. 8, 9, 10 and 11 illustrate the routine

  of control appearing in a general way on the

Figure 6

  Figures 12 to 17 show details of the subprofit

  control and service gram shown generally in FIG. 7, FIGS. 18A, 18B, 18C and 18D show in FIG.

  form of graphs the waveforms of different ano-

  malities that may appear in the signal of electro-

  A patient's cardiogram and their time relationship, and Figures 19A, 19B, and 19C illustrate segments constituting valid R waveforms detected by the control and service routine of the figures.

12 to 17.

  The figures, and particularly Figure 3A, represent

  feel a cardiac control device 10 to be connected by leads 22a, 22b and 22c to a

  patient 20 to receive signals to be processed and in particular

  bind to pack, before storing them in a memory to

  direct access 26 of a portable data entry unit 12

  born. The control device 10 can control the heart of a patient for a prolonged period, for example

  three hours and twenty minutes, while treating and

  continuously receiving the data received in real time for

  memorize them in a way that facilitates reading and

  subsequent display by a display unit 30 in a form

  me dense that the doctor can easily discern. In by-

  In particular, the electrocardiogram signals of a patient are continually monitored to obtain cardiac data in the form of mean trends, maximum frequencies

  and minimums during consecutive intervals, for example

  five minutes, by data entry. In addition, the device stores data in such a form that it is possible to display histograms of R-R interval differences and heart rates. In addition, the cardiac control device is adapted to detect certain arrhythmias and, upon detection thereof, to automatically store for a limited period of time an electrocardiogram signal segment indicative of arrhythmia.

  detected. When the data entry period is complete

  born, the control device 10 is assembled so

  the portable data entry unit 12 is connected to its uni-

  30 for the purpose of triggering a reading and

  display on an analog recorder 36. The unit of res-

  The invention is divided into a first part constituting an adapter 32 and a second part comprising the recorder 36 and a lateral printer 38. As will be explained in detail later, the printer 38 comprises a group of elements which print a series marking on the edge

  recorder document 36 in order to obtain an indica-

  on the occurrence of certain events during the

  control period; for example, the appearance of an ano-

  cardiac malaise detected by the input unit 12 is noted

  by marking which can be placed on a trend curve.

  produced in order to indicate the interval in which this anomaly occurred. Moreover, the patient 20

  when he detects heart pain, presses a button

  marking tone 84 appearing in FIGS. 3B and 4A, so that the cardiac signals appearing during this period are identified by a marking produced by the printer 38. In addition, the printer 38 can print markings to separate the different curves which

  are printed by the recorder 36.

  Figure: A represents in a general way the arrangement

  cardiac control device in which electrical signals are

  trocardiogram are applied by the drivers 22a, 22b

  and 22c from electrodes attached to the patient 20, by the

  a 18-way differential preamplifier

  portable unit 12 and, in particular, to a conver-

  Analog-digital source 16. Signals converted to digital form are applied to the central processing unit

  14. The input of the converter 16 is connected by the amplifier

  differential indicator 18 to> connectors 22a and 22b. The am-

  differential plifier 18 includes a preamplifier

  electrocardiogram which amplifies the signals on

  22a and 22b to reduce the noise produced in the cable. Preamplifier attacks an amplifier

  differential 18 with a passive ground circuit. The connection

  22b is attached to an electrode

  a gel located at approximately V5 of the patient's side,

  say that the connector 22a is attached to a coating electrode

  frost, located approximately on the sternum. These preamplifier connections produce a positive R-wave for most patients. The ground connector 22c is fixed

  on a gel-coated electrode located on the

  left toral. The embodiment shown in FIG. 3A controls the patient's heart from the outside, but it is obvious that the invention also applies to the direct connection of the detection electrodes to the patient's heart and to the transmission of the signals. heart

  detected from a transmitter implanted to an external receiver

  internally connected appropriately to the data input unit; in this case, the transmitter and the receiver-may be in the form described in the patent of

  United States of America No. 4,166,470.

1 1

  A signai. clock from a sample clock

  24 is applied to the central processing unit 14 to control the frequency with which the signals

  samples are taken from patient 20. He is well

  that the central processing unit 14 has its own internal clock, in the well-known manner, for controlling its own operations as well as its interface with the other elements of the unit 12. The central processing unit 14 is connected by a bus line 27 of data and instructions to a permanent memory 28 and a direct access memory 26. The permanent memory 28

  is intended to memorize the instructions that the central unit

  tral 14 executes to detect or recognize the input electrocardiogram signals, to process these signals

  including the settlement and to memorize the signals processed

  in specified positions of the random access memory 26; the programs or subroutines stored in the permanent memory 28 will be described with reference to FIGS. 6 and 7 in a general manner and in more detail with reference to FIGS. 8 to 17. The data thus processed

  and compacted by the central processing unit 14 are

  morris in specified areas of memory access

  According to their nature, the areas of the

  in which the data are stored being

* shown in Figure 5.

  The reproduction unit 30 is optically coupled

  with an isolation flip-flop circuit or power adapter

  32 which behaves as a digital isolation buffer receiving from the data input unit 12

  information to be transmitted and displayed on the record

  analog treater 36 and the edge printer 38. Thanks to the optical isolation, the patient 20 can be connected to the input unit 12 when it is itself connected to the restitution unit 30 powered by the sector without risk that the patient's electrocardiogram signal is affected by parasites or that the subject may experience an electric shock. As shown in Figure 3A, the

  data are applied through a convergent

  The analog-to-analog recorder 34 to the analog recorder 36 may be in the form of a pen recorder manufactured by Astromed Model 102. The toggle circuit 32 is also connected to the printer 38 by

  example that manufactured by Texas Instruments under the name

  EPN 3300. Figure 5 depicts a data arrangement or a

  memory organization showing how the differences

  Data types are stored in corresponding regions of the random access memory 26. As shown in FIG. 5, the random access memory 26 comprises

  a first zone 26a intended to receive data of

  electrocardiogram continuously controlled by the

  10 when operating in the control mode.

  In this mode as will be explained in detail later, each QRS complex of the patient's electrocardiogram signal is detected and analyzed to determine if it is

  valid, ie if there is no parasite and, on the other hand,

  parition of a valid QRS complex signal, the signal is processed and stored in the first zone 2'6a. There occurs

  to keep a series of electrocardiogram signals when

  an arrhythmia is detected which may appear after a desired portion of the signal has already been detected. For this purpose, the first zone 26a of the memory operates

  like a fictitious loop tape by memorizing each

  data to lower and lower addresses of the memory until the lower limit of a section

  from zone 26a is reached. At this moment, the memo point

  P address is returned to the top position of this section, and the new data is written over the old ones. The cycle is repeated continuously until an arrhythmia is detected and at that moment an indicator to the first and last limit of the data

  is memorized and any other data storage is inhibited.

  speechless. When the storage is complete, the indicator is preserved in a starting point position 26a-9 to direct the output of the data. The area 26a of the memory 26 is divided into several buffer sections, for example 9 sections 26a-0 to 26a-8, each section comprising a given number (for example 360) of consecutive positions in which the samples representing the data of a Electrocardiogram signal in four is placed in the above sequence. When a buffer section is filled, the address or indicator P of the zone 26a is moved to indicate the upper available position in the next buffer section. In this way, nine

  series of electrocardiogram signals can be recorded

  stored in memory 26 to render in a manner to be explained, a visual image of the signal to be

  this moment. Although nine sections of buffer are

  in area 26a of memory 26, it is quite obvious

  40 or more sections could be incorporated

  in this area, the only limitation being the capacity

  of memory and its price, and consumer considerations

  current on the battery. In another mode of

  of the invention, a single section of circular buffer

  could be appropriate, since the stored data is co-

  placed in permanent memorizing positions of the

  memory 26 upon detection of a cardiac abnormality.

  The second zone 26b is assigned to the memorization of

  data to be displayed in a historical format.

  gram. In particular, zone 26b is exemplified in the form of two groups of 50 cells at 16.

  bits. The first group of 50 cells is intended to

  an interval difference histogram, in which

  each box has a duration of four

  calibration, ie 16.2 / 3 milliseconds. As will be explained later, when the R-R interval has been calculated and the difference between the successive intervals has been obtained, this difference is divided by four, the

  fractional part being eliminated. The result of the division

  is used to indicate the appropriate cell of the

  interval difference histogram group in the first

  first group in zone 26b, and this cell is incremental

  to indicate the appearance of a difference in this

  special value. In one embodiment of the invention

  each of 25 boxes of difference histogram is provided on the positive side and the negative side of the center.

  to deal with differences ranging from 0 to 100 hor-

  sample box, ie 0 to 41, 2/3 milliseconds

  condes, all differences greater than this value being neglected. The second group of 50 cells in area 26b is for storing the frequency histogram, wherein each box has a width of five beats per minute. As will be explained later, when

  the R-R interval was measured and the cardiac frequency

  that equivalent has been calculated, this frequency is divided

  by five, the fractional part being neglected. The re

  The result of the division is used to indicate the box

  Frequency histogram property and this box is incremented to indicate the occurrence of this particular interval. The interval difference histogram is used primarily to indicate instabilities of

  rhythm that can be produced by palpitations beyond

  ricular, sinus syndrome and contraction

  matured while the frequency histogram indicates the

  distribution of intervals. On the other hand, the trend curve

  dance indicates the large interval variations.

  The zone 26c of the random access memory 26 constitutes

  a storage area to receive data from

  dance, consisting of three groups of 40 cells with eight bits, each cell memorizing or the maximum or the minimum of the

  the patient's heart rate and the average of that

  detected during this interval, for example 5 minutes

  your. Each group of cells corresponds to an interval

trend.

  In addition, the zone 26c makes it possible to memorize data indicating whether the marking button 84 has been operated or

  if an electrocardiogram has been memorized during an inter-

  Trend sampling value. When the sick person

  a symptom, for example dizziness, syncope or other discomfort indicating poor heart condition, it can press the marking button 84 disposed on its portable unit 12 and performed as shown more particularly in Figures 3B and 4B. At the operation of the button 84, a specific bit pattern is stored in the cell corresponding to the trend sampling interval in which the button 84 has been manipulated. If a series of electrocardiograms has been preserved in the area 26c during a given trend sampling interval, the bit pattern of the corresponding cell is changed to indicate this fact. At the restitution, an indication is given on the analog recorder "36 concerning the interval in which the button 84 has been

  maneuvered under the form of a marking identifying the in-

  time, so that the examining doctor can identify the moment when the button is operated or when the electrocardiogram

  the corresponding values of minimum frequency, frequency

  maximum frequency and average frequency.

  Area 26d is a data storage area that receives different variables including indications and status markers that are set during execution

of the process.

  FIGS. 4A, 4B, 4C and 4D are a detailed diagram of the cardiac control apparatus 10 shown generally in FIG. 3A. The data acquisition unit 12 is shown mainly in FIGS. 4A and 4X, which show the central processing unit 14 which can be, for example, in the form of a unit

  manufactured by Intel under the designation 8085A.

  In conjunction with an address register 44, the central unit

  processing gate 14 receives and transmits data by the address 46 bus line, the control bus line 50,

  data bus line 48 and connectors J1 and J2.

  Data is received from the analog-to-digital converter

  16, shown in FIG. 4C, and are transferred to and from the random access memory 26 on the data bus line 48 as shown in FIG. 4A. The memory 26 is constituted by several memory elements such as those manufactured by Harris under the designation Hm 6514. The data bus line 48 is connected to

  Data Transfer terminals DQ1-4 of each of these

  ments. The permanent memory 28 is constituted by two memory elements such as those manufactured by Intel under the designation 2716 which, when addressed, issue program instructions which will be described by the

  following by their 00-07 exit terminals on the omni-

  data bus 48, to the central processing unit 14.

  As shown in FIG. 4A, addresses are applied by an address bus line 46 to the elements of the direct access memory 26 and the permanent memory 28 to address one of the elements and a selected position that they contain. The central processing unit 14 delivers an initial group of addresses to these outputs A10, 11 and 12 to apply them to two decoders of pellet or elements.

  and 42 consisting for example each of a decoder manufacturer

  RCA under the designation 4028. The outputs of the deco-

  40 and 42 provide address signals so that one of the chips or elements of the memory 26 or

  memory 28 is addressed at a given moment. The posi-

  particular in one of the elements of the Memorials 26 and

  That is addressed is selected by a register of addresses

  44 whose input signals come from the terminals ALE and ADO-7 of the central processing unit 14, to output

  the output signals applied by the public address line.

  46 to each of the above to address a

determined position.

  Other control signals are developed from the memory to determine whether the data can be transferred between the random access memory 26 or whether data is to be transferred to the input / output and

  in particular, by the data bus line 48

  analog-to-digital converter 16 in the figure

  4C, or to control the transfer of data to the

  address box 140 shown in Figure 4D which

  constitutes a part of the register 32 represented on the

  Figure 3. The terminals designated by RD and WR control whether data is to be written or read in the random access memory 26. As shown in FIG. 4A, the central processing unit 14 uses a group of interrupts signals. a way which will be explained in more detail later, to simplify the programs which are necessary for the execution of the various operations and to fulfill the functions of recognition and

  memorizing the electrocardiogram data. For example

  4C, the sampling clock 24 applies its sampling signal via the conductor 80, the connectors P1 and J1 to the interrupt input RST 7.5 of the central unit. treatment

  15. The sampling clock calls for a sub-program

  electrocardiogram and time sampling

  tion by which samples of the electronic signal

  cardiogram are taken and the R-wave of the QRS complex is examined to determine if a valid R wave is detected and set up an RSFLAG indicator as will be

  explained with reference to Figures 12 to 17.

  Another interruption is made when a circuit

  voltage sensing device in the form of a floating circuit

  their Schmitt 54 being part of the food source.

  52 shown in FIG. 4A detects that the voltage

  battery dropped below a pre-determined level

  mine. As shown in FIG. 4C, a battery 82 is

  connected by connectors P1 and J1 to the power source

  52 and consists for example of a battery of

  18 volts; when the tilt circuit 54 detects that the voltage

  When the battery 82 has dropped below 10 volts, an output signal is generated and applied to the TRAP interrupt input of the central processing unit 14 to initiate a controlled transfer to a stop mode. operation as will be explained with reference to Figure 7, so that only the addressing and processing logic circuits and memory 26 pads remain powered, thus saving battery power. In

  In particular, the rocker circuit 54 comprises an amplifier

  operational calculator Ul9 which compares the output voltage of

  the battery 82, normally 18 volts, with the regulated voltage

  5SD supplied at output 55 of source 52 and, if the

  battery voltage drops below a value of

  For example, at 10 volts, the operational amplifier U19 delivers a high level output signal through the NOR or U8 gate to the TRAP interrupt input of the unit.

  processing unit 14 in order to trigger the sub-pro-

  604 gram low voltage detector and service,

  as will be explained with reference to FIG.

  The circuit shown in FIG. 4A of the input unit 12 reacts in the following manner to this low voltage condition. First, the terminals A15, IO / M and RD of the central processing unit 14 all go high so that the NAND gate U5 applies a low level output signal to a pulse contender circuit. constituted by a capacitor C16, a diode CR5 and a capacitor R4o. The output signal formatted is used in turn to set the state "1" to

  tipping circuit 57 constituted by two doors

  each signed by U8. As shown in FIG. 4A, the delayed output signal 56 of this flip-flop circuit 57 is applied to the power source 52 by the transistor Q1 as well as to a group of NAND gates U4 and U5. When the transistor U1 is unblocked, a current is applied to the

  switching regulator U20 so that the power source

  tation 52 de-energizes most elements of the

  tif 10, as will be explained later. Of this

  way, the power source 52 removes the power supply

  most of the elements of the device 10 while maintaining the power supply of different elements of the random access memory 26 and further, the group of the doors

  AND U4 and U5 is closed to prevent any application of

  In this way, the data previously stored in the elements of the memory 26 are maintained and, in addition, address signals and / or spurious signals are provided.

  which could be produced by the central processing unit

  14 and / or the decoders 40 and 42 can not be

  bounced to the memory 26 after a power failure.

  Furthermore, the output signal of the flip-flop circuit 57 is used to unblock the transistor Q2 of the power source 52 so that a positive signal is applied to the input SD of the regulator circuit U20 which may for example consist of a regulator manufactured by Silicon General under the designation SG box 24. In the normal mode of operation, the regulator U20 periodically supplies the base of the transistor Q3 with its outputs AC and CB with a frequency that charges the capacitor C14

  at the desired voltage, of the order of +5 volts; the condensation

  C14 is recharged repetitively as it tends to discharge through the circuit of the inductance coil L1, which makes it possible to apply a constant and regulated voltage of +5 volts to the output terminal 55, also designated by + 5SD. In addition, periodic voltages are also applied by the primary winding of the transformer T2 to its secondary windings so that voltages of + 8 volts are applied to the supply of elements of the circuit of Figure 4c. The linear controller U20 delivers at its output VR a regulated voltage of +5 volts. As shown in the review of Figures 4A, 4B, 4C and 4D, most

  elements of the device 10 are powered by the

  controlled voltage + 5SD from the output 55 of the power supply 52 while operating in the normal control mode. But when the transistor Q2 is unblocked upon the occurrence of a low voltage condition, an input signal is applied to the SD input, causing

  switch the U20 controller to the second non-functional mode

  tion in which the + 5SD regulated voltage is removed.

  most of the elements of the device 10, which

  greatly reduces the consumption of the battery 82.

  In this stop mode, a voltage of +5 volts is applied

  only by the VR output terminal of the power source

  tation 52 to the various elements of the memory with

  rect 26, pellet decoders 40 and 42 on the doors

  NO AND U4, U5 and U6 and the tilting circuit 57. In particular

  In this case, the normally applied output signal from the regulator U20 terminals CA and CB goes high to block the transistor Q3, thereby disconnecting the voltage from the primary winding of the transformer T2 and the inductance coil L1, and removing the output voltage

  + 8 volts derived from the secondary winding of the transformer

  T2 meter and the output voltage of terminal 55. But,

  as indicated above, the U20 regulator con-

  continue to apply the voltage of +5 volts from its VR output no

to the aforementioned elements.

  The interrupt input RST 6.5 of the central processing unit

  processing 14 is used to detect whether the preamplifier

  The connector 18 is connected by the connector J4 to the unit

  12 of data capture shown in Figs. 3A, 3B. As shown in Figure 4C, pins 1 and 2

  connector J4 are connected to connector P4 of the pre-

  amplifier which has a short circuit between its pins

  1 and 2 so that a low level signal or 0 is applied by the driver 94 and the connectors P1 and J1 to the interrupt input RST 6.5 of the central processing unit 14. As will be explained next to the

  FIG. 5, in the absence of a connection with the preamplifier.

  18, the device 10 can not enter its mode of -

  control or calibration operation.

  In the control mode or the calibration mode, a

  switch 84 shown in FIG. 4C can be manipulated

  to apply a signal by a conductor 86 and connectors P1 and J1 to the SID input of the central processing unit 14. In the control mode, the maneuver of the

  switch 84 causes the memorization of an identical indicator

  a relatively small interval during the

  the period of control, in which the patient

  detected a certain symptom or felt some discomfort.

  In the calibration mode, the operation of the switch 84 causes an increase in the associated refractory period

  with the detection of the QRS wave. The SOD terminal of the central unit

  In FIG. 4C, the driver transistor 87 and the connectors J1 and J2 conduct the transistor Qi, so that the electro-luminescent diode ECG 88 is energized. As will be explained, during the calibration mode, a potentiometer R9 in the amplifier 19 shown in FIG. 4C is set to establish the gain of this amplifier at an appropriate level as a function of the amplitude of the electrocardiogram signal of the amplifier. sick. The

  electroluminescent diode 88 is excited so as to indi-

  if the R9 potentiometer has been set correctly. A

  start button 96 shown in FIG. 4C is

  operative to initiate operation of the device 10 by which a low level signal is applied to the input

  IN resting of the central processing unit 14.

  In particular, the start button 96 is operated to apply a low level signal through the terminals P1 and

  J1 to a NAND gate U5 whose output signal is applied

  when the tipping circuit 57 is idle.

  is in the "0" state, the rocker circuit 57 applies a

  low level signal that blocks transistor Q2, so

  that the high level signal applied to the SD input of the

  U20 switchgear is removed, allowing the

  to return to normal power mode. In addition,

  the central processing unit 14 delivers an output signal

  which is applied to the quitting and emptying of the address register 140 as shown in FIG.

Figure 4D.

  As shown in Figure 4A, a computer clock

  The transmitter 15 delivers a clock signal at 1.843 MHz which

  timed the execution of the program by the central processing unit 14 as well as the various other data transfer functions ordered. As will be explained later, the functions of the central processing unit 14 include various data transfer and calculation functions that are executed at the control of the computer clock 15 and in its time frame while most data entry functions, including

  electrocardiogram signals are sampled

  folded in real time to the control of the sample clock

24.

  As shown in Figure 4C, the elec-

  input trocardiogram are applied to the current type electrocardiogram pre-amplifier 18. The

  output signal of the preamplifier 18 is applied to the

  electrocardiogram pucker 19 which includes amplifiers

  Operational drivers 90 and 92 connected in series. As has been waved above, the variable potentiometer R9 controls the gain of the electrocardiogram amplifier

  19. The output signal of the latter is applied to the

  The converted digital signal is transmitted by the outputs BO-7 of the analog-to-digital converter 16 via the data bus line 48 to be selectively stored at the control of the central processing unit 14 in the elements. of the direct access memory 26, as the

Figures 4A and 4B.

  Figures 4A and 4E show the isolation register

  32 and the digital-to-analog converter 34,

  shown generally in Figure 3A, connected to the analog recorder 36 and the edge printer 38. The isolation register 32 includes the address register 140, a sixteen channel optical isolator 141, and a DC-DC converter 139. The data to

  print or display are applied by the omni-

  data bus 48 at the inputs D0 to D7 of the address register.

  140. The latter issue a first group of

  PBO to PB7 which are applied by the isolate

  16-channel optical transmitter, shown in Figure 4D,

  and the digital-to-analog converter 34, as the

  Figure 4E, so as to attack the analogue recorder

logic 36.

  The register 140 delivers a second group of output signals PAO to PA7 which are applied to the printer 38 to

  seven points. In particular, the output signals of the re-

  140 are applied to a group of U4 optical insulators UIL to two channels, as shown in Figure 4D. The insulators U4 to Uil are provided to isolate the analog recorder 36 and the on-board printer 38 which can be powered by an ordinary power outlet of the unit 12.

  data entry to ensure that signals, including

  The supply voltage received by the recorder 36 and the printer 38 are not transmitted to distort the electrocardiogram signals from the patient or put him at risk of electric shock. The signals of

  isolated output from the optical isolators and corres-

  the output signals PA1 to PA7 and PBO to PB7 are applied by connectors P4 to the digital converter 34 and to a group of drive transistors Q1 to

  Q8 respectively. As shown in more detail in the

  gure 4E, the collectors of the drive transistors Q1 to Q8

  are connected to the inputs of the connector 35 of the printer.

  FIG. 4E does not explicitly represent the edge printer 38, but it is understood that a group of printing elements in the form of heating resistors is incorporated in the analog recorder 36, these resistors being fed to print selectively one or more

  markings a1Gn to indicate the different curves or

  data. In particular, the analogue recorder

  36 can make appear a presentation of a histogram

  R-R in a manner similar to that

* shown in Figure 1, a histogram of difference of inter-

  similar to Figure 2 and a trend sample showing a series of sampled intervals of consecutive trends, for example 5 minutes, of the minimum and maximum heart rates. During the presentation by the analog recorder 36, the printer 38 produces a series of markings on the same paper to separate the

  different curves and, as far as the sample curve is concerned,

  trends, it produces a series of markings

  different to indicate the occurrence of different

  24. in a sample of trends. For example, a

  threefold markings above a sample of indi-

  that the device has detected an anomaly, as

  Ra explained in more detail. If the patient detects a possibility of discomfort related to his cardiac activity and if he presses the marking button 84 in response to this detection, the edge printer 38 is energized to place a double mark near the test sample. trend. Yes

  an excessive parasite is superimposed on the electro-

  cardiogram from the patient 20, the appearance of this parasite is indicated by the printer 38 which forms a

  single marking above the trend sample cor-

  sponding. Through conductors, the connector 35 supplies the heating resistor elements, not shown, for

form the aforementioned markings.

  As indicated above, the register of addresses

  wise 140 applies a series of PBO output signals to PB7

  optical isolators U8 to Ull to the digital converter

  Analogical signal 34 which delivers an output signal I to an operational amplifier U2. In turn, the output signal of the operational amplifier U2 is applied to the signal input of the recorder 36 so that its

  feather moves to represent the analog signal from

  Tree. The circuit shown in FIG. 4E comprises a device in the form of a potentiometer R36 through which

  an offset can be applied to the signal received by the

analog recorder 36.

  As will be explained in more detail later,

  the particular mode of operation in which the

  Tif 10 is dependent on the elements of the device that are connected to the data input unit 12, as shown in Figure 3A. For example, the device operates in

  in the calibration mode if the unit of re-

  30 is connected to the portable unit 12. As shown in FIG. 4D, the addressing register 140 has outputs PC1 and PCO which are placed respectively at the high level and at the low level by the signals coming from the outputs PC5 and PC4. If the connector P3 is connected to the connector J3 as shown in FIG. 4C, the access can be given to the address register 140 to determine the

  signals applied to the PC1 and PCO terminals, so that the

  positive 10 enters its standard operating mode.

  swimming or restitution. In addition, input signals are output to outputs A11 and A12 of the central processing unit.

  processing 14, as shown in FIG. 4A, towards the

  Ai and AO of the address register 140 to determine whether the particular signal applied by the data bus line 48 is to be transmitted to the recorder 36 by the outputs PBO to PB7 or to the edge printer 38 by the

outputs PAl to PA7.

  Figure 6 is a general flowchart of the se-

  wherein the cardiac data control device according to the invention operates to record and tamp

  cardiac data and to view the stored data.

  in a form that intelligibly informs the doctor of the state of the patient's heart. Initially,

  phase 200, the operator presses the button 96 to start

  rage and put to rest on the control panel of

  the data capture unit 12 shown in FIG. 3B.

  Then, in phase 202 interruptions of the central unit

  Trale treatment 14 are inhibited. As was ex-

  above, the central processing unit 14

  has different interrupt inputs that allow it to

  attempt to receive data from different areas of the permanent memory 28. As will be explained with reference to FIG. 7, the interruption allows automation

  by the circuits of different operations in order to simplify

  Prove the software and save the execution time. initially

  In addition, interrupts of the central processing unit are inhibited. Interruptions are allowed by a series of markers that are arranged in a matrix or

  an interruption register of the central unit 14, allowing

  both to selected interrupt inputs to be used

  while inhibiting others. Initially, each of the interrupt inputs of the central processing unit

  is inhibited and, subsequently, an initialization operation

  allows selected interruptions by applying

  2a 73874 markers to the corresponding areas of the interrupt matrix. Also during initialization, the signals are applied by the central processing unit 14 to position the isolation register 32, to allow the flow of data to and from the analog recorder 36 and the on-board printer. 38 and to detect whether the rendering unit 30 is connected to the input unit 12. Next, the selection phase 204 is executed during which one of the following modes of operation is selected: restitution (PBKM), calibration (CALIBM) and control (MNTRM). For the most part, phase 204 reviews

  the configuration of the whole with regard to

  which are connected to the unit of sase 12, to determine

  in which mode the device 10 should operate. For example, if only the rendering unit 30 is connected to the input unit 12, the phase 204 transfers the program to the output mode by the phase 300. But if the reproduction unit 30 is disconnected from the unit 12 and if the preamplifier 18, in particular the connector J4 shown in Figure 4C is connected to the portable unit 12, the device 10 goes into the control mode 500 by the initialization phase 400. In particular, the J4 connector which interconnects the preamplifier 18 and the input unit 12 has two pins connected together, so that the input RST 6.5 of the central processing unit is short-circuited or brought to the low level, so that a input state is stored in a unit register

  central office, to which access is given during the selection phase

  In a similar manner, the connection of the input unit 12 to the reproduction unit 30 and in particular to the isolation flip-flop circuit 32 is via the input / output terminals of the central unit. 14 who reacts if the

  Isolation switch circuit 32 is connected to it.

  Before entering the control mode 500, it is necessary to control the authorization marker indicating that the device has been calibrated in the calibration mode 206. Otherwise, if the authorization marker is not detected,

  the device can not enter the control mode 500.

  The MCLEAR 400 subroutine returns the marker of

  after directing the program to Control Mode 500 to avoid entering this mode again

  unless calibration mode 206 is performed again

  calf to set the mopper to "1". To enter the calibration mode 206, it is necessary to detect that the input unit 12 is connected to the preamplifier 18,

  by detecting short-circuited pins at its connection

  J4, as well as to the isolation flip-flop circuit 32. Furthermore, if the input unit 12 is not connected to any other device, the device immediately goes out to standby mode 208. In the calibration mode 206, the CALIBM (Calibration Mode Monitor) program is run to assist the operator in adjusting the device to receive different amplitudes of heart signals and for

  regulate different refractory periods which are

  necessary to correctly detect the QRS complexes of a patient. With regard to the amplitude of the R-wave of a patient, it can vary in a ratio of 20: 1 in a normal population. So, to avoid saturation

  of the analog-to-digital converter 16 and for

  to operate in a range of 30 to 50% of its total capacity, it is necessary to adjust the gain of the input amplifier 90 of Figure 4C forming part of the converter 16 until the normal amplitude of the R wave is at a level that allows a 2: 1 increase in the electrocardiogram signal amplitude without reaching the upper and lower countdown limits

  of the converter. In this mode, the output of the preamplifier

  The driver 18 is continuously monitored through the converter 16 to determine whether the positive or negative peak-to-baseline deviation exceeds 40 units or count of the converter. If the peak input is below this limit, the CALIBM program does not excite the ECG88 light-emitting diode shown in FIG. 4C. If the peak amplitudes produce peak deviations from the baseline exceeding a second limit corresponding to 60 units of the converter 16, the diode 88 is lit continuously. If the peak gap from the baseline measured by the converter 16 is between 40 and units, the light emitting diode 88 flickers whenever a QRS complex is detected. Normally, the gain of the amplifier 90 should be adjusted so that the output of the converter 16 around the baseline deviates by about +50 counts during the cardiac cycle. For this purpose, the variable gain of the input amplifier 90 of the converter 16 is set by R9 so that the diode 88 flickers on each QRS complex. In addition, the calibration mode 206 regulates the refractory period, that is to say

  the expected period between the R and T waves of the cardiac signal.

  than the patient, during which electrocardiographic signals

  gram applied to the input unit 12 are received but not

  processed to detect QRS complexes. To adjust the

  refractory period, the marking button 84 is pushed in order to

  a high-frequency hum or signal is applied

  only on the electrocardiogram curve for a period that lasts as long as the button 84 is pushed. This signal is developed by the analog recorder 30 and its

  duration increases every four seconds while the operation

  The operator holds the button 84 until the refractory period indicated by the snoring reaches the end of the T wave.

  will be explained later, the refractory period is cal-

  in accordance with the time adjustment subroutine

  fractary, as a function of the duration of the last RR interval and of a constant that is modified only during the refractory period setting, ie the corresponding counter of the refractory period counter in turn sets the constant calibration that is used to

  calculate the refractory period on the successive appearances

sives of the R-R interval.

  After calibration, indicated by the establishment of the

  authorization and detection that the data entry unit

  12 is connected to the preamplifier 18, the selection phase

  204 makes a connection to the quit and initialization subroutine 400, executing the MCLEAR program in which the memory areas indicated on the

  Figure 5 are emptied to receive the refractive constants

  The latter will be explained later with reference to FIGS. 8, 9, 10 and 11 and is controlled by the MNMRI program to detect successive QRS complexes, to determine if these complexes are valid or not valid for characterizing valid QRS complexes, for measuring RR intervals and for controlling the treatment of these

  data and the storage of the data processed in the

  Direct access memory 26. The R-R interval is calculated and the corresponding frequency is obtained before storage

  in zone 26c of memory 26, as indicated by the

  Then, the differences between the successively calculated RR intervals are calculated to increment one of the boxes in the area 26b of the memory 26. In addition, a classification routine is provided for detecting abnormal or arrhythmic cardiac signals. ; in particular, it is determined whether the frequency is greater than 120 or less than 40 beats per minute or whether there has been a skipped or premature beat. If so, the marker

  preservation is in place and a warning signal

  is sent to memorize the current group or the current series of electrocardiogram rhythm signals in one of the nine buffers of the zone 26a of the memory 26. FIG.

  long after the marker of

  the buffer data are immobilized and the indica-

  to the memory buffer address moves

  to the next buffer if available, in field 26a.

  The control subroutine 500 controls the operations

  in the control mode. It does not contain any sub-

  real-time task execution program such as interval detections, R-wave detections and R-R interval calculation. Real-time subprograms are separated, as shown in Figure 6, to simplify the overall software in the device. In summary, the real-time tasks including the R-wave detection routine are introduced once during each cycle of the sampling clock 24. The control routine 500 is

  executed as long as the march marker remains in state "1".

  n0 Any of the following events can bring it back to "0"

  ending the control mode: first, the fight against

  The power supply of the device 10 goes below a predetermined level; then, the area 26c which receives the trend data can be filled, i.e. no other data can be stored; and third,

  the connector J4 of the preamplifier 18 can be disconnected.

  When the runner is set to "0" the program MNTRM performs the final management tasks of phase 500 and transfers the command to phase 208 in which the standby mode monitor de-energizes all elements

  of the whole except for memory 26. In particular,

  link, at the end of each of the operations of the subroutines 300, 206, 500, the operations go to the standby mode 208

  in which interruptions of the central processing unit

  14 are inhibited and the data stored in the

  direct access memory 26 are protected against

  ture or writing. In addition, the power source 52 is disconnected from all parts of the device 10, except for the supply of the random access memory 26, thus ensuring that the data stored therein

  are kept safely. In this way, the consumption

  battery 82 is reduced to a minimum, thus ensuring

its maximum longevity.

  In addition, the selection phase 204 can directly transfer the process to the standby mode 208 if the

  preamplifier terminal 18 as well as the reproduction unit

  30 are disconnected from the data input unit 12.

  Moreover, and as shown generally in FIG. 6, a real-time interface program 600 is provided to organize the different tasks necessary to execute in real time, for example the detection of the R wave, in the form of a service for fashion monitor programs that can be run in computer time. These real-time service routines may detect certain conditions directly placing the device 10 in the standby mode 208. Figure 7 illustrates in more detail the real-time interface program 600. Phase 602 detects whether the output voltage of battery 318 is low and, if so, it controls the unit.

  processing center 14 so that it enters a sub-

  program 604 or low voltage detection service, so that the process is transferred to the standby mode 208 As shown in FIG. 4A, a Schmitt swing circuit 54 is connected to detect the output voltage of the battery 82 In response to detecting a voltage below a predetermined value, for example 10 volts, the circuit 54 applies a high level signal or +5 volts to the TRAP input of the central processing unit 14.

  the latter should move to subprogram 604 of the

  Low voltage detection fault. In this program, it is first determined whether operation is in control mode 500 and, if so, the walk marker is set to "O". As will be explained in detail later with reference to FIG. 8, the control mode 500 responds to the "0" setting of the march marker, assuring

  that the electrocardiogram data are completely

  registered, that the calculations for the current R-R interval are

  completed and the average of the interval obtained is calculated

  and memorized in an appropriate position of the memory at

  direct access 26. Then the latter is blocked from

  not to lose the stored data and to maintain

  its feeding, as described later. The cir-

  tilt 54 controls the TRAP interrupt of the central unit

  14 to interrupt the program in the standby mode 208, in order to reduce the programming that would otherwise be necessary. The sentence

  206 detects the absence of the preamplifier 18 and in particular

  link, that the terminal J4 associated with this preamplifier is disconnected, that is to say that the short-circuited pins are removed from the interrupt input RST 6.5 of the unit 14 and this interruption input passes at +5 volts so

  that the central processing unit 14 triggers a sub-pro-

  preamplifier state serving program 608 in which the process can be transferred to the standby mode 208. In particular, if the device is operating in the control mode 500, the service routine 608 passes it in a loop until the preamplifier 18 is connected again to the input unit 12 or until the marking button 84 is pushed. When the button 84 is operated, the counter marker is set to "0" so that the program MNTRM puts the device into the standby mode 208. While the device is in the loop, no other data is available. is stored in memory 26 or processed to detect QRS complexes. Moreover, when the marking button 84 is pushed, the average

  R-R signals received at that time are calculated

  memories stored in different other positions of the memory 26 are erased. If the operation is done in

  the calibration mode 206, the device transfers directly

  in standby mode 208 upon detection of disconnection

  of the preamplifier 18 of electrocardiogram signals.

  The present embodiment has a single button

  84, but it is clear that a series of these

  could be included in the device, so that the patient can specify more specifically the type of sensation he has perceived as well as a particular symptom.

  a body function or activity. In addition, a provision

  an accelerometer type fixed on the body of the patient

  would replace the marking button 84 to indicate by a rapid movement of the body, a high rate of activity the

  contact that can be closed so that the electronic signal

  cardiogram be recorded during a given period.

  As shown in FIGS. 4A and 4C, the clock of

  Sampling 24 delivers a sample clock signal-

  240 Hz at the RST 7.5 interrupt input of the CPU 14. The sampling clock

  controls the central processing unit 14 so that it can

  makes a 612 crystal clock subroutine so that

  the appropriate mode service subroutine is introduced

  as shown in Figure 7. In particular, the sub-

  program 612 controls the current mode of operation, ie determines which one of the modes 300, 206 or 500 is in progress and, according to this mode, it selects one of the service routines 614, 616, 618, 620 or 622 to execute. In particular, subroutine 612 examines an XCMODE mode state variable assigned to a specified position of memory 26 to determine whether its value is 0,1,2,3 or 4 corresponding to subroutine 614, 616, 618 , 620 or 622 to be selected. Subprograms 614, 616, 618, 620 and 622 are executed in real time and provide an interface between the signals

  which are received in real time and the other

  calculations and calculations that are carried out by the monitors of

  corresponding mode in the calculator time to the command

  of the clock 15 connected to the central processing unit

14.

  A calibration subroutine 614 performs the basic real-time synchronization for the monitor in real time

  calibration mode CALIBM 206. In particular, the sub-

  program 614 responds to the sampling clock in

  repeatedly quoting the wave detection subroutine

  R so that the QRS complex of the electrocardiographic signal

  gram is detected and its amplitude is established. The

  electrocardiogram peak signals obtained are pre-

  and are transferred to the calibration mode monitor 206 which, in response to the computer clock 15, determines

  if the amplitudes of the QRS complex are acceptable. In-

  In this case, the calibration routine 114 applies the electrocardiogram signal and the buzzer signal as required to the analog recorder 36 to facilitate the recording.

  refractory period setting as explained above.

  above. The control subroutine 616 behaves like a

  real-time interface for the control mode 500, rem-

  folding the basic synchronization functions, soliciting

  both the R wave detecting subroutine and the memorization of one of four electrocardiogram multiplets, ie of its corresponding amplitude, in one of the nine rotating buffers constituted in the zone 26a of the direct access memory 26. If, during the

  control 500, a marker is set to "1" by the sub-

  MNTRMI program indicating the detection of an arythmi-

  that the current buffer of zone 26a is preserved and the

2473874-

  subroutine control 616 produces a marker for the

  next buffer, if available, in field 26a.

  For the most part, the sub-program of curves

  618 ensures the synchronization of the different curves, so that data is transmitted from the memory to

  direct access 26 at a rate corresponding to the progress

  Step by step the paper entered by the recorder 36.

  For example, each cell of trend data indi-

  when minimum and maximum average intervals R-R are applied to the recorder 36 at such a speed that

  cell is visualized by millimeter in advance of the paper.

  It should be noted in this regard that paper advances by one

  meter by 40 milliseconds. To get this synchronization

  tion, subroutine 618 responds to the output signal of

  the sampling clock 24 by producing a clock signal

  drive bay for the recorder 36. It should be noted that

  that the sampling clock can not be sim-

  only 40 milliseconds, but that a synchronization subprogram is provided to react to 10, 10, 9, 10, 9 clock

  corresponding advance signals for the recorder 36.

  The 48 clock signals appearing in a range of

  milliseconds accurately equivalent to five inter

  Values of 40 milliseconds. The fact that each advance signal

  slightly erroneous does not affect the overall accuracy

  the curve drawn by the recorder 36.

  The data curve subroutine 618 operates in real time to perform a rendering of the trend and histogram data stored in the memory.

  on the analog recorder 36 and the onboard printer 38.

  In addition, subroutine 618 provides a rectangular signal

  60 Hz to the analog recorder 36 to form markings between the curves it traces. For example, an interval is marked between the trend line and the histogram curve to make it clear to the doctor

a division between these curves.

  The subroutines 620 and 622 for rendering and positive and negative slope respectively react to the positive and negative slopes of the electrocardiogram signal received by interpolating between electrocardiogram points so as to form a series of rectilinear segments between

  two neighboring points and produce a new appearance curve.

  on the recorder 36. A series of

  lines, each 16.67 milliseconds in length, is used to reproduce a curve of regular appearance between the two points. If it is noted that the data acquisition unit 12 is a portable unit and contains a random access memory 26 whose capacity is limited, it is desirable not to memorize all the samples.

  signals from the patient's electrocardiogram to increase

  the duration during which the control device 10 can operate. But, at the restitution, when the data stored in the memory 26 are read, it is necessary to interpolate between the points stored in the memory

  26 to obtain a regular appearance curve.

  The control mode 500 described and shown generally in FIG. 5 is shown in greater detail in FIG. 8. The selection phase 204 transfers the control to MNTRiM 502, the input point of the control mode monitor 500. through the MCLEAR program

  400. The order then moves to phase 504 to introduce

  to warm up and to memorize variations in

  For example, it is necessary to position the indicator to specify an initial buffer position in the zone 26a, in the trend sample zone 26c and in the zone 26b of the trend sample.

  of histogram 26b. In addition, all memo positions

  in the minimum frequency region of the trend data area 26c are set to a maximum value to simplify the necessary programming steps

  to determine the minimum frequency of the heart beats

  diaques. In addition, a clock count variable

  sample, to determine a sample of elec-

  trocardiogram in four is memorized to allow

  to count a number determined by the successful appearances

  cessives of the sampling clock, and counting down

  to zero the electrocardiogram sample corres-

21.73874

  is stored in the specified position of zone 26a. A variable called XCMiODE, placed in the memory 26,

  is set to "1" to indicate in step 612 that the

  tif works in the control mode. Thus, the sub-pro-

  gram 616 of the control service may be required for

  provide the control mode monitor with an appropriate service

  requested in real time, for example synchronization and QRS detection, at each positive transition of the sampling clock. A count to determine the duration of the trend interval according to a determined number of sampling clock signals is intended to be

  counted on the occurrence of each sample clock signal.

  to determine, for example, a five-minute interval at the "O" count. In addition, a blocking time is stored to introduce a delay before a marker is set by MNTRM to perform the memorization

  permanent data in a buffer of zone 26a.

* After memorizing these data as well as others, -

  the process goes from the loop input 506 to the decision phase 508 which determines whether the march marker has been set to "1" or "0". As indicated above, the march marker is set to "0" if it is determined that the voltage of the battery is low, if the connector J4 of the amplifier 18 is disconnected, if the zone 26c of

  storing of the trend or the area 26b of history storage

  tography is full or if the indicator that specifies the

  pon in zone 26a indicates that all buffers are already

  full. If the running indicator has been set to "O", the

  The gram passes through step 520 in which the sample data or the current trend interval is calculated to obtain an average based on the R-R intervals already received. Thereafter, the control is transferred to the standby mode monitor 208 in which commands are issued to the power source to power off all elements of the device except memory 26 and the associated chip selection circuitry to maintain storage. of all the data already registered with

  reduced consumption on the battery.

  If on the other hand in phase 508 the march marker is

  to "1", ie if the device is in working condition

  In order to receive, process and store electrocardiogram signals, the process proceeds to step 510 to check whether the marker button 84 has been operated and, if so, the step 512 stores an event marker. in an auxiliary matrix of past behavior, in the memory 26, indicating the sample or the interval in which the button 84 has been manipulated. Then, after phase 512 or if decision phase 510 indicates that the tag button has not been manipulated, the process proceeds to phase 514 to determine if the tag R has been set to "1" by

  execution of the R-wave detecting part of the sub-pro-

  control service program as will be described in

  Figure 16. If this is the case, the operations

  Subprogram 700 will be described in detail hereinafter with reference to FIG. 10. In the subprogram 700, the data established by the execution of the control service routine 616 shown in FIG. , and in

  particular the R-wave detecting subroutine

  are used to reset the marker R to "0" to process the data of the R wave type indicating the presence of parasitic noises stored in a specified area of the memory 26 and to process the RR interval stored in the memory. one of the buffers of the memory 26. After the execution of the subroutine 700 or the decision phase 714 indicating the

  marker R to "0", the process moves to phase 518 in the-

  the elapsed time in a trend sample interval is compared with the end value of stored TSI

  in zone 26d to determine whether the exchange interval

  trend lilting is past, for example if five minutes are expired. The elapsed time is stored in a counter formed in the zone 26d of the memory 26 and is incremented

  at the appearance of the sampling clock. If the

  valle is passed, that is to say if the answer to phase 518

  is positive, a subroutine 800 which will be explained in re-

  Fig. 9 summarizes the data stored in the

  completed trend sampling to determine

  undermine the R-R range moven and therefore the average frequency of the beats and he then directs the memorization

2A73874

  R-R interval data to the next position or the next cell of the area 26b. If decision phase 518 indicated that the trend sample interval

  has not expired or at the end of subprogramme

  gram goes to phase 516 in which operations are interrupted until the sampling clock

  following appears and results in the execution of the sub-program

  control service which introduces and processes new

  electrocardiogram data from the patient.

  Then, the process ends its loop, returning to the point

  506 to examine and process new electronic data.

  cardiogram. The recapitulation routine 800, represented by

  generally in FIG. 8, is shown in more detail in FIG.

  Fig. 9 shows that this routine is entered in phase 804 to initiate summarization operations. In particular, at phase 806, the

  sum of the R-R intervals in a single sampling interval

  trending, memorized in a memory position

  of zone 26d, is divided by a count indicating the

  number of R-R intervals detected during the same

  Trend sampling value. Then, the average interval is inverted to obtain the average frequency of

  heartbeat of the patient during the sampling interval

  trend sampling. Then, in phase 808, the indica-

  TSI PTR is incremented by one so that the data

  R-R interval corresponding to the sample interval-

  Next trend scan is stored in the next cell or slot of the memory 26c. Then, at the phase

  810, the current TSI PTR indicator is compared with an indi-

  the maximum capacity of zone 26c and, if the capacity

  maximum city is exceeded, the march marker is set to "0" in phase 812 so that, as explained

  compared to Figure 8, phase 808 transfers the function

  step 820 to standby monitor 208

  If, at the determination of phase 810, there still exists -

  of the memory available in the zone 26e or if the march marker has been set to "0" by the phase 812, the phase 814 erases the counters formed in the zone 26d of the memory

  26 and which are used to count the number of

  and add them during the new current trend sampling interval. Then, in phase 816, the end indicator of the

  current trend, is reduced to "10" for indi-

  the end of the interval by adding 300 seconds

  condes to the current value of the elapsed time counter.

  Phase 818 then brings the process back to the sub-program.

  MNTBM control as shown in Figure 8.

  The processing and storage of the R-R data

  in general by program 700 of the

  Figure 8, appears in more detail in Figure 10. The sub-

  program enters phase 704, as shown in Figure

  10, for processing and storing data concerning the in-

  R-R associated with the QRS complex recently detected by the 616 control subprogram. In phase 606, the

  marker R is set to "0" to acknowledge receipt of the

  detection of a QRS complex by subprogram 616. In addition,

  phase 606 obtains the variable of the R wave type, indi-

  as to whether the detected QRS complex is contaminated by

  sites, and an R-R interval representing the measured R-R interval. Then the R-R interval value is placed in

  the position of the memory 26d called new position R-R.

  Then, the R-R combination value is made equal to the current R-R sum with the new R-R value, thus adding up the two most recently recorded values.

interval R-R.

  As will be explained in detail later, the new RR values corresponding to the most recently calculated R-R interval and the R-R combination are used to determine whether to preserve a series or

  a group of electrocardiogram signals from the patient

  corresponding to the detection of an anomaly, as will be explained with reference to FIG. 11. It is necessary to classify R-wave data so that the intervals associated with valid R-wave signals are treated by neglecting signals contaminated by parasites. In addition, intervals following a premature beat 40 must be treated in a different manner than those following a regular rhythm. As shown in Figure 10,

  a phase 708 is planned to evaluate a type of wave R va-

  riable. The R-wave type change of value strategy, in conjunction with phase 708, is the means of classifying each detected QRS complex. For example, it will be assumed that the QRS complexes are not contaminated with parasites and that the device has just been turned on, ie the R wave type has been set to " 4 ". Therefore, when the very first QRS complex is detected, phase 708 transfers control to phase 728 which does not use any R-R interval information because

  this information is meaningless when no

  anterior valle was not detected. Since the interval associated with the next QRS complex will have significance, phase 728 sets the R wave type to "3". Therefore, the next uncontaminated complex is treated in phase 724

  in which the detection of the first non-contagious QRS complex

  This is used to initialize the device and memorize the following uncontaminated signals. In phase 724,

  series positions in which the most

  of the RR interval, namely the new RR interval, the current RR interval, the previous RR interval and the penultimate RR interval are shifted to place the most recent RR interval value in the position new interval of memory. Then the program

  establishment of refractory duration is requested to establish

  Blind the refractory period based on the most recently detected interval between QRS complexes. Next, the R wave type is set to "2" to process the QRS complex

  following that can be considered normal, uncontaminated.

  After executing phase 724, program 700 returns

  by phase 726 to phase 518 of control program 500.

  Upon detection of the next QRS complex, indicated by the transition to "1" of the marker signal R, the process returns to subroutine 700, in which, since the wave type R has been set to "2" step 708 directs the program to phase 722 in which an acceptable QRS complex is stored as a regular or premature rhythm signal. The refractory duration setting routine is executed to establish the refractory duration according to

  of the R-R interval and then the correction subroutine

  difference is required to calculate and memorize the difference between successive R-R intervals in

  the difference histogram of zone 26b of memory 26.

  Next, a ranking and storage routine shown in Fig. 11 is performed to determine whether the detected QRS complex without contamination is part of a regular lorythmus or premature beat. if

  part of a regular rhythm, the R wave type is left

  se a "2" whereas if the QRS complex appears premature

  the waveform R is set to "1". The regular rhythm and the premature beats are treated and memorized as will be explained with reference to FIG.

  different positions to receive and memorize the inter-

  R-R valleys are corrected with the most recent value of

  the interval R-R stored in the new position

  valle. After the execution of phase 722, program block 20700 exits from phase 726 to phase 518 of the program.

control 500.

  Upon detection of the following QRS complex after the transition to "1" of the R wave type, at the detection of the "1" state, the phase 708 transfers the control to the phase 720 in which the 251st QRS complex that has been preceded by a premature beat is treated. This beat is usually known as a compensation beat and it is not memorized in

  trend zone 26c of memory 26. But the difference

  this between successive R-R intervals is calculated and the difference correction routine is solicited

  to memorize the calculated difference in the group of his-

  Difference togram of zone 26b. In addition, the period

  refractory is established by the establishment subprogramme

  refractory period depending on the last interval

  351st QRS calculated. Then, the R wave type is set to "2" to condition the acceptance routine to receive the next normal QRS complex. In addition, the areas of the memory 26 intended to receive the previously calculated and stored values of the R-R interval

  are corrected with the most recent value of this

  stored in the new interval position.

  In the case where the service service subroutine 616

  truce determines that the signal is contaminated by

  sites, this subroutine sets the wave type R to "0" so that the phase 708 transfers the command to the phase 718 in which the RR interval that has just been measured is not stored because it is contaminated. But the wave type

  R is set to 2 "to allow the initialization of the

next QRS plex, uncontaminated.

  The acceptance subroutine 722, represented by

  General in Figure 10 is described in detail with reference to

  He has the function of dealing with the

  electrocardiogram-of the patient and memorize data that has been considered a valid and acceptable R wave. The acceptance subroutine 722 begins.

  phase 730, in which the subprogramme

  refractory period is requested to establish the

  refractory period in which the device 10 is inhi-

  to detect any other activity or cardiac signals on the input leads during this period, after

  the appearance of the R-wave. In general, the refractory period

  of a patient whose heart rate is 75 bat-

  per minute is in the order of 300 to 400 milliseconds.

  The refractory period is variably established by the refractory period setting subprogram, based on the previously determined RR interval stored in the new interval position, in a rotating buffer of the memory area 26d. it

  has been explained above, this rotating stamp comprises four

  positions in which four most recently determined consecutive values of the R-R interval are

  stored in order in the positions of new

  valle, current interval, previous interval, and penultimate interval. When a new value of the R-R interval is calculated or selected in the same way

  explained in relation to figure 16 it is first memorized

  in the new interval position of the buffer pre-

  quoted, and the old values of the R-R intervals are

2 '738 74

  transferred to the following positions. The subprogramme

  refractory duration selects the value of

  the R-R interval placed in the new interval position

  and, using the following expression: refractory = KQT (new interval +27/60), it determines the duration of the

  refractory period in the form of sample clock counting

  from the clock circuit 24. Then, at step 730, the difference correction routine

  asked to determine the difference between

  neighboring R-R valves by subtracting the value stored in the previous interval position from that stored in the current interval position. The calculated value of the difference between neighboring R-R intervals is used to set an indicator to a cell or cell in the interval difference histogram of the zone

  26b of the memory 26, and this cell is then incremented

  one unit to count the occurrence of this difference.

  between the current interval and the previous interval.

  Decision phase 732 is executed to determine whether a premature beat has occurred based on the most recent data indicating the RR intervals stored in the new interval, current interval, previous interval, and before positions. last interval of the above-mentioned rotary buffer. In particular, there are two

  types of premature beats that are detected, a bat-

  premature with a compensation pause and an interpolated beat; these types of beats are detected in that they can be attributed to malfunctions of the ventricles and they are considered more serious

  premature beats resulting from a bad functioning

  the atrium. The appearance of a pre-

  matured with a compensation pause is illustrated in Fig. 18A which shows a series of R, RO, R1, R2, R3 and

  R4, the RO wave appearing first and the R4 wave the last one

  Niere. As shown in the figure, the R3 wave is prematurely shifted to the right in Figure 18A, to the R2 wave; in addition, the space between the R3 wave and the R4 wave is greater than normal, establishing a compensation pause, so that the R4 wave appears at about the expected time

  normally. An intervening premature beat is repre-

  FIG. 18C on which an R3 wave is

  approximately halfway between the beats corresponding to

  R4 and R2 waves, the interval between the R4 and R2 waves being approximately equal to the normal R-R interval. As will be explained, subroutine 732 detects whether it is a premature beat with compensation pause or

  premature beat intercalated. Batte-

  prematurely are treated and memorized from a first

  by phases 736 to 742 while if no bat-

  premature birth has occurred, the R-R interval is treated

  in a second way to phases 748 to 764.

  Determining whether a beat is premature or not is done by examining the values stored in the new-interval, current interval positions,

  interval and penultimate interval of the

  turn, and in the following manner. Subprogram 732 determines whether a first condition is met, that is

  whether the interval R-R stored in the positiôn of in-

  current period is less than 75% of the value

  rised in the previous interval position; this com

  parison can be viewed by examining Figure 18B

  on which the interval between the waves R2 and R3 is in-

  less than 75% of the interval between the waves R1 and R2.

  Then, to determine the existence of a break from

  sation, the interval combination value indicating the sum of the R-R intervals stored in the new interval and current interval positions, is compared

  with a value of 175 Io of the memory R-R interval

  se in the previous interval position. As shown

  18B, the distance between the waves R4 and R2 is

  trimmed with the length between the waves R1 and R2 and if it exceeds 1.75 times this last distance, and if the first condition is fulfilled, the decision phase 732 gives an indication that the beat is premature with a pause of compensation, by transferring the order to the

phase 736.

  Second and third tests are carried out to de-

  terminate if an intervening premature R3 beat exists,

2 '; 73874

  as shown in Figure 18C. In particular, it is deter-

  mined if the R-R combination value indicating the sum

  values in the positions of new interval and

  The current interval is less than 1.25 times the value of the RR interval stored in the pre-set interval position.

  yield. As shown in Figure 18D, this indication

  the total length between the wave R2 and the wave R4 with the length between the waves R1 and R2. Next, a third test is performed to determine if the value stored in the R-R combination position is less than 1.25

  times the value of the R-R interval stored in the posi-

  penultimate interval. In other words, the distance between the waves R2 and R4 is compared with the distance between

  the RO and Rl waves to obtain an indication of the stability

  frequency of heart beat; if the frequency is very irregular, the detected beat may not be

  a premature beat but an indication of another ano-

  more dangerous malice. If the first condition and the

  second and third tests are positive,

  decision-making procedure 732 that a premature beat

  Interleaved rate has been detected and the program is

phase 736.

  The phase 736 shown in Figure 11 is the beginning

  of a pre-term beat treatment routine.

  Phase 736 sets the R wave type indicator to "1", so that the processed R wave is considered an anomaly

  as a result, the memory storing routine 700

  R-R in Figure 10 is executed again, then

  a connection is made to the subroutine 720 of

  anomaly in which only interval

  the R-R are memorized, and the device is prepared for

  receive the next R wave as an acceptable R wave. In-

  Next, phase 738 examines the past behavioral position in the area 26d of the memory 26 in which

  counted and memorized a number of premature beats

  res in the current trend sampling interval,

  in order to determine where there is room for another

  crémentation. If space is available, the position of

  past behavior is incremented by one unit in phase 740.

  In any case, the program proceeds to phase 742 in which the waveform storage control routine is biased to trigger the operations performed by the control routine that preserves a waveform sample. containing the premature beat in one of the flow buffers of the zone 26a of the memory 26. In addition to triggering the waveform storage, the phase 742 of the memory storage control

  waveform introduces a memory blocking system.

  which avoids the filling of the zone 26a of the

  in a few minutes by a series of pre-programmed beats

  matured or other arrhythmia. In particular, when the order

  of waveform storage is requested to memorize

  signal, it compares the current value of time spent

  with the type of arrhythmia to preserve. If the blocking period has not expired, the storage lock command

  waveform simply returns without triggering any

  sample. If the blocking period has expired a sample preservation is triggered and the blocking period is triggered to prohibit memorization

  other similar arrhythmias for ten minutes, add

  both 600 seconds to the current period of time elapsed

  to determine when memorizing an exchange of

  This particular arrhythmia signal will be allowed.

  Then, at step 744, the positions of the rotating buffer progress so that the R-R intervals stored in the different indicated positions are placed in

  the next position, and the last calculated value of

  R-R is in the new position of

valle.

  If the decision phase 732 determines that the beat

  or the examined R-wave is not premature, the program does not

  Phase 748 is the first step in a non-premature beats treatment routine. Phase 748 calculates the current frequency by inverting the value of the stored RR interval in the current interval position and furthermore, the cells of the current box of the trend sample area 26c are corrected with respect to regards the trends of maximum frequency and

2 ', 738 74

  minimal. In particular, the cell positions in which frequency trends are stored

  maximum and minimum are compared with the current value

  Calculated heart rate and if it is higher or lower, the corresponding position of the box is corrected with the new maximum or minimum frequency. The determination of the maximum and minimum frequencies gives a statistical indication of the regularity of the heartbeat and, as will be explained, it may be displayed for the convenience of the physician. It should also be noted that according to the invention, the differences between the successive intervals of the cardiac activity signal

  calculated to obtain the histogram of different

  valles can also be used in calculations to determine the standard deviations of the different intervals and that these standard deviations can be determined for each trend sampling interval and be

* - stored in an appropriate memory position corresponding to

  laying at this interval. Then, at phase 750, a

  part of the recently chosen or calculated value of the

  R-R stored in the current interval position is added to the sum position R-R in zone 26d

  26. The R-R interval, unlike the use of

  the total value, is used to reduce the

  memory space needed in the sum position R-R.

  As indicated above, at the end of an interval

  the particulars of trend sampling, the sub-pro-

  TSI summing gram is requested to calculate the inter-

  R-R mean value by dividing the sum of the intervals R-R stored in the sum position R-R by the number of

samples taken.

  Then, as shown in Figure 11, it is determined

  if an abnormality other than a premature beat has occurred, and particularly if it is a tachycardia, bradycardia or a skipped beat, in order to set up a series of markers and for that particular sample the electrocardiogram signal corresponding to the anomaly is stored in one of the circulation buffers of the zone

  26a of memory 26. First, in phase 752, the

24.73874

  their most recently calculated heart rate

  is compared with the maximum frequency, above the-

  what an indication is given about the presence of an ano-

  malice in the heart of the patient known as tachycardia.

  In addition, phase 752 determines whether the frequency is increasing by comparing the values of the stored R-R intervals.

  in the current and new interval positions

  value and, if the value stored in the new interval position is equal to or greater than that stored in the current interval position, indicating that the frequency is beginning to decrease, the program proceeds to phase 754 in which the routine waveform storage control is required to set up the markers by which a series of electrocardiogram samples indicating the tachycardia are placed in a rotating buffer of

  the area 26a of the memory 26. For example, the frequency

  The selected maximum is set at 120 beats per minute. If the most recent heart rate value is not greater than this maximum level, the routine proceeds to phase 758 in which the most recently calculated frequency is compared with a minimum frequency indicative of the occurrence of bradycardia. and, if it is less than the minimum, for example 40 beats per minute, phase 760 requests the waveform storage control routine to set up markers by which a series of electrocardiogram data illustrating bradycardia is stored in a rotating buffer of the area 26a of the memory. If it is not less than the minimum value indicated, the process proceeds to phase 762 in which it is determined whether a beat has been skipped. In particular, the value of interval RR stored in the current interval position is compared with 1.75 times the value of the RR interval stored in the previous interval position and, if it is higher, it

  indicates a beat skipped so the process goes on -

  at step 764 to request the waveform storage control program to set up the

  markers by which the series of electronically

  cardiogram of the patient is stored in one of the buffers

U'73874

  in circulation of the area 26a of the memory 26. After each

  phase 754, 760, 762 and 764, phase 756 incremental

  The 744 corrects the values of RR intervals in the positions of the rotating buffer, before returning to the MNTRM program and in particular to step 518 of FIG. 616 shown generally in FIG. 7 will now be described in FIG.

  details in Figure 12. First, the sub-

  800 ECG sampling and synchronization program

  is requested in response to the detection of the hor-

  sampling box at the input of the CPU 14. As will be explained in more detail in

  FIG. 13, a sample of the electronic signal

  cardiogram is taken and processed before being memorized.

  Four of these samples are stored and their average

  is done before placing them in a position of multi-

  ECG plet. In addition, subroutine 800 determines whether an R wave is present and, if so, sets it to "1".

  the marker R and further determines whether the R-wave is conta-

  mined or not by parasites; if the wave is contaminated by parasites, the wave type R is set to "0". The subprogram 800 represented in a general way on the

  Figure 12 will be explained in more detail with reference to

  13 to 17. In phase 802, it is determined whether the

  third sample of electrocardiogram was taken in

  checking the clock count and if it is "0", the sample

  the corresponding one is the fourth. In addition, phase 802 determines whether there is room for storing the next sample in one of the nine rotating buffers in area 26a of memory 26. If it is the fourth

  sample and if there is space available, the program

  go to phase 804; otherwise it returns to the point of

  tie 820 to the main program. In phase 804, the

  indication of the average of the last four samples

  is stored in a position of one of the buffers cir-

  culation identified by the PTR loop indicator. Then, this indicator is incremented to select the position

2'73874

  next in the rotating buffer. Then in phase 806, the program examines whether the ECG preservation marker has been set to "1" by the waveform storage monitor

  requested in one of the phases 742, 754, 760 or 764 for indi-

  respectively, an abnormality such as tachycardia, bradycardia, or a skipped beat. If the marker of

  ECG preservation is at "1", the program goes to phase.

  808 in which the loop remainder counter is decompo-

  to allow a given number, for example 60 samples

  additional tones of the electrocardiogram signal is stored in the rotating buffer after detection of an anomaly. In this regard, it should be noted that a total of 360 electrocardiogram samples are arranged in one of the

  rotating buffers in area 26a of memory 26. If the

  in phase 806 is negative, subprogram 616

returns to the main program.

  Phase 810 determines whether the swab was filled with electrocardiogram data, delaying the execution of the other phases until the 60 samples of the signals

  electrocardiogram of the patient appearing after the

  of the ECG preservation marker have been introduced,

  what is decided in phase 810. If this is the case, the

  gram goes to step 812 in which the start and end flag of the filled swivel is memorized

  to allow access when it is necessary to read the

  stored in this buffer. In addition, phase 812 places

  the memorisation indicator on the first positinn of the drum.

  next available circular pon in area 26a of the

  26. At phase 814, it was decided whether there was another

  3rotation buffer available in zone 26a and if not,

  the march marker is set to "0" so that the

  gram MNIRM 500 is directed on the transfer of the command to the waiting subprogram 208. Stil-exists another available buffer, which decides the phase 814, the phase 816

  loads the loop remain position with the pre-counting

  determined additional samples of electrocardiographic

  gram, for example 60, which must be memorized after

  detection of the anomaly and then, the indicator of

  ECG service is set to "0" to complete the preservation

CE '7387

  samples in one of the rotating buffers. Then, the program returns through the output 820 to the execution of the phase of the control mode monitor program 500 which

  was interrupted by the sampling clock.

  As has been indicated above with regard to the

  7, the control service subroutine is requested

  cited in response to the appearance of each of the hor-

  sampling box if the device 10 of FIGS. 3A, 3B operates in its control mode, ie if the preamplifier 18 is connected to the data acquisition unit 12. Subprogram 800 ECG sampling and

  synchronization system takes samples of the signal of

  electrocardiogram in response to the sampling clock signal.

  the 240 Hz frequency, and determines whether a valid R wave is present, like this

  will be explained in more detail later with regard to

  13 to 17. As regards Figure 13, the sub-

  program 800 begins with a synchronization phase 822 in which the light-emitting diode 88 shown in FIG. 4C is excited for a fixed period, for example 40 milliseconds, corresponding to 24 counts of the sampling clock, in response to the detection of each wave R; in particular, a count 24 is loaded into the diode delay position of the direct access memory and is decremented until it is

  equal to "0", in which case an order is issued to disqualify

  In addition, the phase 822 advances two counters in the area 26d of the memory 26 and which

  serve.to synchronize two separate functions. All of

  edge, the elapsed time position is incremented each

  second and it's used as a timer for

  seems to be triggered at the beginning of the control or calibration mode; the elapsed time variable is used to control the timing of the trend sampling and, in addition, the timing of the

  blocking during which other series of elec-

  trocardiogram can not be memorized as explained above with respect to subprogram 742 of

2 73874.

  waveform storage control of Figure 11. The

  second clock is constituted by the position of temporisa-

  RR is used to synchronize the interval

  R-R, ie the interval between valid R waves.

  The triggering and interruption of the temporary position

  RR will be explained later. Then, at the

  se 824, an ECG input and filtering subroutine is

  solicited by which a sample of the electro-

  Diogram is taken and stored in a four-position rotating buffer which is incremented each time a new sample is obtained. Electrocardiogram data

  are filtered after a sample has been taken in

  killing the average of the contents of the four-position buffer and

  memorizing the result in the ECG multiplet position.

  At the onset of the next sample, the model data

  ECG yen stored in ECG multiplet positiont is

  transferred to the previous ECG multiplet position, the

  turn is corrected and a new average is stored

  in the ECG multiplet. The ECG byte and the preceding ECG byte are used, in a manner to be explained later to determine if a valid R wave has been detected. Then, in phase 826, the program examines the current contents of the RR timer to determine if the

  refractory period has elapsed and, in particular,

  counters the current count of the RR timer with the refractory period determined by the refractory period setting routine, as explained above. If the refractory period has elapsed, as determined by phase 826, the segment search routine 900 will be

  described in more detail in relation to Figure 14 is requested

  cited to determine the existence of a R wave segment.

  After execution of subprogram 800 or if it has been indicated

  that the refractory period has not yet elapsed, as determined by phase 826, phase 828 makes a decision

  indicating whether a possible R wave segment has been found.

  As will be explained in detail with respect to Fig. 14 the segment search subprogram places the segment status position at "1" or "2" if a valid segment has been

2 73874

  found or at "3" if a short segment was found; so, if

  phase 830 finds a "1", a "2" or a "3" in the posi-

  segment status, the command goes to the subprogram

  Segment analysis in which segments detect

  are compared to determine if a valid R wave has been detected. When subroutine 1000 has been executed or if

  the segment state was found equal to "0", the process not-

  phase 830 in which the state position of seg-

  is set to "0" to prepare the processing of the sample.

  following electrocardiogram. Thereafter, the program exits through step 832 to terminate either the control service routine 616 or the calibration service routine 614. In particular, if the device is in the control mode, the program returns to

  phase 832 to phase 802 as shown in figure 12.

  The segment search subprogram 900 represents

  Figure 13 will be explained in more detail with reference to Figure 14; in this subprogram,

  phase 902 initially evaluates the mode position of

  search for memory 26 to determine the state of the sub-pro-

  900 grams; the search mode position is placed ini-

  tially to "0" by the initialization phase 54 of the sub-

  control program 500. When the search mode is

  placed at "0", the program moves to phase 906 to start

  the operation of detecting the beginning of a potential wave segment R. As shown in Figure 19, the QRS complex in

  which the R wave forms a positive peak, can be

  approximated by a series of segments. Figures 19A and B show examples of valid R waves in

  which the electrocardiogram signals have been

  felt like a series of three segments. As will be

  explained below, the existence of a valid R wave is detected

  by measuring the time interval between segments, the length, slope and amplitude of these segments and their number. Initially, in phase 906, the interval time or space between two consecutive segments is measured and, if it is greater than 62.5 milliseconds, it is decided that already detected segments are part of a QRS complex or a noise signal because the maximum interval between two associated segments has been exceeded. Therefore, the program moves to phase 916 in which the position

  segment status is set to "3" to indicate that the

  maximum span between two valid segments has been exceeded.

  If the maximum allowed space between two segments is not exceeded, as decided by phase 906, phase 908 stores the current data in a segment start position of memory area 26d and initiates synchronization of a period of time. corresponding to the maximum duration of a valid segment, for example 95.8 milliseconds. Current data

  include the previous value of the electrocardiographic signal

  average gram stored in the ECG multiplet position pre-

  yield, and the present value of the time from the posi-

  RR timer. The next phase 910 determines if there is a change in slope and if this change of. slope

  is positive, comparing the value stored in the posi-

  ECG multiplet and the previous ECG multiplet position

  tooth. If the first value is greater than one second of a predetermined value, indicating a positive slope greater than a threshold value, the phase 912 sets the seek mode position to "1" to direct the segment search routine 1000. , as shown in Figure 16, so that it looks for the end of this segment. If the decision is negative in phase 910, phase 914 determines whether a negative slope segment is present by examining the ECG byte and the preceding ECG byte and, if the value of the preceding ECG byte is greater than the value stored in the position. ECG byte of a predetermined amount, this indicates that a start of segment has been found, and the process proceeds to step 918 in which the search mode marker is set to "2" to direct the routine 1000 segment search to find the end of this segment. After any of the phases 916, 912 or 918, the program returns to phase 828 of subroutine 800

  ECG sampling and synchronization.

  At the next prompt of Segment Search Routing 900 and if phase 912 or phase 918 set the search mode marker to "1" or "2", phase 902 detects the particular marker state and directs the process to phase 922 or 928 to locate the end of a segment. For example, it will be assumed that the beginning of a positive slope segment has been detected at a previous pass by the segment search 900, so that the search mode marker has been set to "1". Therefore, evaluation phase 902 transfers control to phase 922 to locate the end of the positive slope segment. First, phase 922 checks for a reverse slope or if the segment duration limit that was set at phase 908 is reached. If there has been an inversion

  slope or if the duration of a segment has been exceeded, the

  gram proceeds to phase 924 in which the amplitude of the segment is calculated in its absolute value and then the routine -e management and end of segment which will be described in detail with respect to FIG. different counters or timers are

  initialized again to prepare for the recall of the subpro-

  gram 900 of oath search so that he finds a new

  calf segment. In particular, phase 924 is reminiscent of

  their previous ECG byte which has been memorized

  908 at the beginning of the segment detection

  and subtracts this value from the previous ECG byte

  the detection of the end of the segment to obtain

  Absolute amplitude of the beginning to the end of the electrocardiogram sample. When phase 924 is completed or phase 922 has answered both questions negatively, segment search routine 900 returns to

  processing of the service phase 828.

  During a subsequent execution of the segment search routine 900 in which the search mode is set to "2", the phase 902 directs the routine to phase 928 in which the end of a negative slope is

  determined in a manner similar to that described above.

  In summary, phase 928 determines whether slope reversal has occurred by examining the previous ECG byte and ECG byte positions of the rotating buffer and / or whether the segment duration limit is exceeded. If either of these conditions is fulfilled, the program proceeds to phase 930 in which the absolute value of the magnitude of the segment is determined before requesting the end-of-segment and management routine which will be explained in detail. FIG. 15. Thus, in segment search subprogram 900, as explained above with reference to FIG. 14, the beginning and the end of a segment as well as its amplitude are detected in preparation for the execution of

  subprogram 1000 segment analysis in which several

  of these segments are examined to determine if the particular combination of these segments thus detected

constitutes a valid R wave.

  Fig. 15 shows the end routine 930 of

  segment and management in which, after the calculation of the

  In step 924 or 930, the various counters and the different storage positions are initialized in preparation for the identification and measurement of the parameters of the next segment. In particular, the phase

  932 compares the measured values of duration and ampli-

  the detected segment with predetermined minimum values

  for example 15 units or counts of the numerical digital converter 34 and 16.7 milliseconds corresponding to four counts of the sampling clock 24. If these minimum conditions are exceeded, the program proceeds to phase 934 in which the value of the mode. search is transferred to the segment state to be used from the

  explained later in the execution of the sub-

  1000 segment analysis program. If the minimum conditions

  males are not filled, what decides the phase 932, it is determined if the maximum interval between segments, for example 62,5 milliseconds is exceeded since the end 3 of the last segment; in particular, the phase 938 compares the current value of the RR timer with the segment start baseline and, if it is greater, the program goes to the phase 940 in which the segment state is set to "3", indicating an elapsed time without valid segment detected. After phases 934 and 940 and if phase 938 responds in the negative, the subroutine goes to phase

  936 in which the search mode marker is placed

  "0" to prepare the segment search routine 900 for detecting the next segment

Z4 738 74

  and in particular, the beginning of this segment.

  As shown in Figures 19A and 19B, a series of

  segments can represent valid R waves. In part-

  In accordance with the invention, the QRS complex is defined by three segments which have a particular slope, amplitude, duration and sequence. As shown by a

  In general, FIG. 13, subroutine 900 of FIG.

  segment search identifies an individual valid segment by obtaining values of its duration, slope, and amplitude, and passes its values to the subroutine

  1000 segment analysis which stores them in posi-

  specific features, examines the sequence of segments,

  class into a QRS signal or parasite, calculates the interval

  the R-R and restores the timer RR. The execution of segment analysis subprogram 1000 will be explained in more detail with reference to FIG.

  phase 1002 access is given to the analysis mode to de-

  end the value of "0", "1" or "2" found

  currently in this position. In the initialization phase

  504, as shown in Figure 8, the method of analysis

  is set to "0" so that initially the

  step 1006 in which the first segment is obtained and analyzed as will be explained. Upon receipt of data corresponding to the first segment, phase 1006

  examine the segment status position to determine if

  the is less than "3". As has been explained about

  of phase 934 of Figure 15, the segment state is

  "1" or "2X corresponding to the detection of valid segments with positive and negative slopes;

  between segments is exceeded, I seg-

  is set to "" by phase 916 or 940. So, if the segment was found valid by segment search,

  phase 1006 stores the segment data in the first

  first segment segment of the memory 26. In particular, the first segment data includes its amplitude,

  the moment it began to be recorded by the time

  RR, its peak amplitude stored in the posi-

  the previous ECG byte and the end of the 58-segment recorded in the timer RR. In addition, phase 1008 changes the scan mode value from "0" to "1"

  so that at the next execution of phase 1002, the

  1000 segment analysis program goes to phase 1012 to evaluate the second segment. Firstly, the phase 1012 determines whether the state is equal to "3" ie if a period exceeding the maximum period corresponding to 62.5 milliseconds has been exceeded since the end of the first segment and, s Thus, phase 1014 requests the parasite processing subprogram 1034, as will be explained later with reference to FIG. 17, to bring the different memory positions to rest.

  timers and counters to erase correct data.

  respond to the first segment and prepare the segment analysis

  to receive and analyze the data of the next segment as for a first segment. When phase 1012 detects

  segment state equal to "3", it indicates that only one segment

  as shown in Figure 19C, has been detected but that a

  second segment did not appear within the time limit

  ptable; therefore, no valid R wave has been detected

  But if phase 1012 indicates that the state of the second segment

  is "1" or "2", ie it is not "3", the

  The program moves to phase 1016 in which it is

  mined if the first and second segments have the same slope; in particular, the phase 1016 compares the segment state positions of the first and second segments and if they are the same, ie if the two segments have the same slope, the program goes to the phase 1018 in which the Data corresponding to the first and second segments are merged together, assuming that only one large segment has been interrupted by parasites or a cardiac conductivity anomaly and only one segment has been detected. In particular, the starting point of the portion of the first segment is retained while the end point of the portion of the second segment is used as the end point of the first combined segment. The amplitude of the first combined segment is taken as the sum of the already determined amplitudes of the first and second parts. In addition, the analysis mode marker is left at "1" so that

2-; 73874

  the new execution of the segment analysis subroutine 1000, the second segment is received and analyzed. On the other hand, if phase 1016 determines that the slopes of the first and

  second segments are different, the identifica-

  the second segment, namely the start and end points and the amplitude are stored in the position of the memory 26d of g and segment data; then, the analysis mode marker is set to "2" to prepare the segment analysis routine 1000 to receive and

  analyze data corresponding to the third segment.

* At the third execution of subroutine 1000 of

  segment lysis with the "2" analysis mode, phase 1002 transfers subroutine 1000 to phase 1022 so that the data corresponding to the third segment can be

  received and analyzed. Phase 1022 compares the slope of the seg-

  in progress with the slope of the previous segment. If it is determined that the slopes of the successive segment are the same,

  phase 1024 assumes that the second segment has been interrupted

  pu and the data corresponding to the current segment are fu-

  with the data of the second segment, that is to say that the starting point is taken as that of the second segment the end point is taken as that of the next segment or

  current, and the amplitude is obtained by adding the

  plitudes of the old and present segment parts. The

  phase 1004 does not change the mode of analysis of its

  "2" and the program then returns to the continuation of the service subprogram at step 830. If the slopes differ, phase 1026 stores the current data including the start and end points and the amplitude in the position. 26d of the memory corresponding to the data of the third segment. Then the analysis mode is set to

  "0" to prepare the segment analysis subroutine 1000 to receive and analyze the next segment sequence.

  Phase 1028 then triggers an assessment of the

  of the three segments to determine whether a valid or invalid R wave has been detected. First, phase 1028 determines whether the amplitude of the first segment is greater than that of the third segment by comparing the

2_473874

  amplitude values stored in the assigned positions for the first and second segments. If the first value is greater, as shown in FIG. 19A, the first two segments, ie from point 1 to point 2 and from point 3 to point 4, are recognized as forming the contour

  of the R wave, then the R-R interval is calculated, the temporality

  RR is reset and the QS duration is calculated.

  In this respect, it is understood that the identification data of the three segments are obtained at a time which follows the

  Detection and analysis of segments; therefore, the

  lysis is done by determining the appearance of the peak of the R wave and, in the first case shown in Figure 19AK

  the crest appears at the end of the first segment and the interval

  the R-R is determined by examining the value of the timer RR at the end of the first segment. In a similar way, the QS duration is taken as the difference between the beginning of the first

  segment and the end of the second. Moreover, he is now born

  restore the RR timer to

  start counting the next R-R interval. For this purpose, the timer RR is brought back not to zero but over a time corresponding to the interval between the end of the first segment and the end of the third. On the other hand, if the first segment is not greater than the third, which corresponds to the second case of FIG. 19B, the last two segments, ie the second one starting at point 3 and ending at point 4 and the third segment starting at point 5 and ending at point 6 are chosen as the contour of the wave

  R, the R-R interval and the QS duration are set accordingly.

  quence and the timer RR is brought to rest. In part-

  In this case, the peak of the R wave is selected at the end of the second segment to thereby define the RR interval as the value of the RR timer at the end of the second segment. In addition, the QS duration is defined as the difference between RR timer values at the end of the third segment and the beginning of the second. The timer RR is set to start counting at the beginning of the third segment; more particularly, the timer RR is brought back to the end of the third segment over a time which corresponds to the duration

2 * 7 3 8 7 4

  the third segment to reflect the real time at rest of the timer, so that the next R-R interval can be measured. After each of the phases 1030 and 1032,

  another check is carried out by phase 1033 to de-

  terminate if the measured R wave passes other validity tests; in particular, it is determined whether the QS duration is

  less than 221 milliseconds and if the R-R interval de-

  completed is greater than 242 milliseconds or 58 counts of the sampling clock. In the context of the invention

  the mean R-R interval determined for the sampling interval

  Trend trending can be compared with a given cutoff period to identify the existence of a cardiac abnormality. Within these limits, as determined by phase 1033, phase 1036 passes the R marker

  to "1" to indicate the detection of a valid R wave. In-

  As a result, the ECG light emitting diode is lit for 83.3 milliseconds by placing the diode timer over a corresponding period. The process continues

  with phase 830 of the service routine. If the

  meters of the detected R wave do not fall within the bounds

  phase 1032, the 1034 phase solicits the

  parasite treatment program, which will be explained in

look at Figure 17.

  If the three segments do not appear in a

  given value, which determines the phase 1012 or if the duration

  QS and the R-R interval are not within the predefined limits

  terminated at step 1032, the noise processing routine 1034 is solicited, as shown in Figure 17. First, at step 1038, the timer RR is

  to "0" and the analysis mode is set to "0" to pre-

  parse the segment analysis subroutine 1000 to receive

  see and analyze data corresponding to a first

  segment of another complex. Then phase 1040 determines

  mine if the R wave type is greater than 2 "and if so, the subprogram sets the R wave type to" 4 "

  initialization during the subprogram of

  storing R-R interval data shown in Figure 8, and in more detail in Figure 10. If the R-wave

2.73874

  is less than or equal to 21 ", indicating an initialization completed, the wave type R is set to" 0 "to ensure

  that the next R-R interval is neglected because it is conta-

  mined by parasites. After phase 1048 or 1042, phase 1044 determines whether the device is in the control mode by checking the XCMODE tag and, if so, the light emitting diode 88 flickers for 41.6 milliseconds to give an indication to the operator that a parasite has been detected; in this respect, this flicker duration of the diode 88 is about half of the

  excitation period of the diode 88 in response to the detection

  a valid R wave. If the device is not in the control mode, which phase 1044 determines, it is

  brought back immediately to the continuation of the execution of the

  flicker-free segment analysis program -

electroluminescent diode 88.

Claims (34)

  1 - Tester and memory device,   for a prolonged period of time, electrical signals   feeling a cardiac activity of a patient, a device   characterized in that it comprises a memory (26) comprising first and second groups (26a, 26b) of storage positions, a device (12) connected to receive   a signal of cardiac activity of the patient, in order to sample   periodically, in real time, enter the   heart activity, and to memorize signals   samples in respective positions of said first   first group, and a data processing device (14) for processing the sampled signals stored in positions of said first group, and storing in a data processing time frame the processed signals in the storage positions of said first group; second group.   2 - Apparatus according to claim 1, characterized   it further comprises a sample clock   (24) and a computer clock (14), said   positive sample responsive to the output signal of said sampling clock by sampling the cardiac activity signals of the patient, said device   (14) for processing data responsive to the output signal.   of said computer clock by memorizing the sample signals ..   3 - Apparatus according to claim 2, characterized   in that it further comprises a sensor (18) associated   with the sampling clock and reacting to a function   identification of the patient's body by producing in time   real a manifestation of his appearance.   4 - Apparatus according to claim 3, characterized in that said memory (26) comprises a third group of storage positions (26c), said data processing device (14) responsive to said event by transferring a series of sampled signals from said   positions of said first group towards said positions of said third group.   Apparatus according to claim 3, characterized in that said sensor comprises a switch (84)   manually operated by the patient in response to a symp- felt.   6 - Apparatus according to claim 2, characterized   in that said data processing device (14)   memory stores the processed signals in the positions of the   second group so that they can be read easily is lying.   7 - Apparatus according to claim 6, characterized   in that it comprises a display device   (30), said data processing device (14) comprises   taking a device for reading data in the posi-   second group so that they are   easily on said display device.   8 - Apparatus according to claim 2, characterized in that said device (14) for data processing   includes a real-time processing device that reacts   said sample clock (24) by processing the stored data in the storage positions said first group.   9 - Apparatus according to claim 8, characterized in that said device (14) for processing in real time   comprises a device for comparing the sampled   stored in the positions of said first group   with validity criteria to determine whether the Samples sampled are valid.   Apparatus according to claim 9, characterized in that said memory (26) comprises a third group   Memory position (26c), said communication device   parison evaluating the sampled signals to determine   undermine whether these signals constitute a valid segment of a   the patient's heartbeat, to give   tation, said data processing device (14) com-   bearing in addition a device reacting to the   a valid segment by memorizing   representing them in a corresponding position of said third group, each position of said third group receiving and storing the identifying data a segment.   Apparatus according to claim 10, characterized   in that said comparison device (14) evaluates the segment identification data stored in the positions of said third group to determine whether   a detected and treated combination of constituent segments   kills a valid heartbeat of the patient, and if he   so is, providing a second manifestation.   12 - Apparatus according to claim 11, characterized   characterized in that said device (14) for processing   data comprises a device for accessing data   storing segments stored in the positions of said third group, and determining the intervals between the valid heartbeat.   13 - Apparatus according to claim 12, characterized   characterized in that said memory (26) comprises a fourth   group (26d) of storage positions, each posi-   said fourth group receiving and memorizing   data indicating the interval between heart beats   successive diaques, said data processing device (14) comprising a device responsive to the output signal of the computer clock by examining neighboring positions of said fourth group to determine whether a valid heartbeat occurs with a regular rhythm and, if not, producing a third manifestation a cardiac abnormality.   14 - Apparatus according to claim 13, characterized   in that said examination device (14) determines   if the valid heartbeat signal appears before   maturely to give the third indi-   when a premature heart beat.   Apparatus according to claim 13, characterized   in that said examination device (14) determines whether the valid heartbeat signal appears at   outside of an expected range, to give the third   - the second manifestation of heartbeat late or jumped. 2/473874   16 - Apparatus according to claim 13, characterized   in that said examination device (14) compares the determined interval with its maximum value and, if   this value is exceeded, delivers a fourth festation of a cardiac anomaly.   17 - Apparatus according to claim 16, characterized   in that said examination device (14) compares the interval with its predetermined minimum value and, if it is lower, delivers a fifth manifestation cardiac abnormality.   Apparatus according to claim 18, characterized   in that it further comprises a device for detecting   reaction to a function of the patient's body in   providing a sixth manifestation.   19 - Apparatus according to claim 18, characterized   in that said device (14) for processing data   The invention includes a device responsive to each of said third, fourth, fifth or sixth manifestations by performing permanent storage of heart activity sampling signals in the positions   storing said first group (26a).   Apparatus according to claim 13, characterized   in that said memory (26) comprises a fifth   group of memory positions, each position of-   said fifth group receiving data indicating the   cardiac activity of the patient, said device (14) for treating   data comprising a device responsive to the   output signal of said computer clock in memory   risking cardiac activity data in packed format,   facilitating reading and display.   21 - Apparatus according to claim 20, characterized   in that the said control period consists in   consecutive intervals of sampling of   each position of said fifth group storing heart activity data corresponding to one of a plurality of consecutive sampling intervals   trend, said device (14) for processing data   with an addressing device that selects   and stores the cardiac activity data in the   current situation of the fifth group.   22 - Apparatus according to claim 20, characterized   in that each position of said fifth group corresponds to a range of a particular characteristic of the cardiac activity data, said device (14)   data processing apparatus comprising a device for   wise of a particular position of said fifth group according to the particular detected range of the data cardiac activity.   Apparatus according to claim 22, characterized   in that said characteristic contains an indica-   the frequency of the patient's cardiac   24 - Apparatus according to claim 22, characterized   in that said characteristic comprises the difference   between the intervals associated with cardiac beats successive issues.   Apparatus according to claim 20, characterized   in that the current addressed position of said fourth group (26d) receives the currently determined interval   between successive heartbeats, calculated   as soon as the previously determined interval signal is erased. Apparatus according to claim 25, characterized in that the previously sampled signal is erased from its position of said first group (26a) when the signal   most recently sampled is stored in the posi-   currently addressed from the first group (26a),   said data processing device (14) comprising   a device responsive to the detection of a car-   by permanently memorizing a series of   samples of cardiac activity in the posi- first group.   27 - Tester and memory device,   during a period of prolonged control, of   indicating a cardiac activity of a patient,   the same characterized in that it comprises a memory 24738? 4   (26) comprising first and second groups (26a, 26b)   storage positions, a sample clock-   generating (24) generating a clock signal stream, a data processing device (14) comprising a unit. central processing that runs programs   to receive, process and store the activity signals   patient's heart, said treatment device   comprising a device responsive to each hor-   sample box by sampling the signals of   cardiac activity of the patient and by memorizing the   in corresponding positions of said first   group (26a), and a device (14) for assessing the   sampled cardiac activity patterns stored in the positions of said first group with respect to   given criteria and, if they correspond, treating and   storing the sampled cardiac activity signals in a format packed at positions of said second group (26b). 28 - Apparatus according to claim 27, characterized in that it comprises a computer clock producing a second train of computer clock signals, said   evaluation device (14) responsive to the hor- decalculator box.   29 - Apparatus according to claim 27, characterized in that it comprises a display device (30), said data processing device (14) comprising   a device for reading data in the positions of   said second group so that they are easily   displayed on said display device.   Apparatus according to claim 28, characterized in that said data processing device (14)   includes a real-time processing device that reacts   said sampling clock (24) by processing data stored in the storage positions said first group (26a).   31 - Apparatus according to claim 30, characterized in that said memory (26) comprises a third group (26c) of storage positions, said device 2XL73874   real-time processing comprising said evaluation device   evaluation of sampled cardiac activity signals   to determine whether these signals constitute a voluntary segment   heartbeat to provide a first manifestation, and a device responsive to the first   manifestation of a valid segment by memorizing   samples representing him in a correct position.   of said third group, each position of that   third group receiving and memorizing data tifying a segment.   32 - Apparatus according to claim 31, characterized   characterized in that said evaluation device (14) evaluates the segment identification data stored in the positions of said third group (26c) to determine   if a detected and treated combination of the constituent segments   kills a heartbeat of the patient's activity and, if   so it is, producing a second manifestation.   33 - Apparatus according to claim 32, characterized   in that said data processing device (14)   has an access device to the identification data.   segments stored in the positions of said third group, and determining the intervals between valid heartbeat.   34 - Apparatus according to claim 33, characterized   characterized in that said memory (26) has a fourth storage position group (26d), each position of said fourth group receiving and storing data indicating the interval between successive valid heart beats, said memory processing device (14) data comprising a signal responsive device   computer clock by examining the positions   of the fourth group to determine whether a bat-   valid heartbeat appears in a regular rhythm and if not, producing a third manifestation to indicate a cardiac abnormality.   Apparatus according to claim 34, characterized   in that said examination device (14) compares the   determined at its maximum value and, if it is 24 73874   exceeded, produces a fourth manifestation of an ano- cardiac malaise.   36 - Apparatus according to claim 35, characterized   in that said examination device (14) compares the   with its minimum predetermined value, and if it is lower, produces a -fifth manifestation a cardiac abnormality.   37 - Apparatus according to claim 35, characterized   in that it includes a device (14) for   an average value of a group of the predefined intervals   completed, said examination device comparing the value   average value with a predetermined value and, if it is   which produces a manifestation of a car- Diaque.   38 - Apparatus according to claim 37, characterized   in that said examination device (14) compares the   their average with a predetermined maximum value, and if it is higher, producing a manifestation of a cardiac anomaly.   39 - Apparatus according to claim 36, characterized   in that said data processing device (14)   has a device responsive to one of said thirds   s, fourth or fifth manifestations by permanently storing heart activity signal signals in the memory positions said first group (26a).   Apparatus according to claim 34, characterized   in that said memory (26) has a fifth group of storage positions, each position of said fifth group receiving data indicating the cardiac activity of the patient and said data processing device (14) comprising a signal responsive device   computer clock by storing the activity data   heart rate in a packed format for easy reading and the display.   41 - Apparatus according to claim 40, characterized   in that the said control period consists in   consecutive intervals of sampling of   each of the positions of said fifth group storing heart activity data corresponding to one of   several consecutive sampling intervals of   said data processing device (14)   taking an addressing device which has cardiac activity data in the current positions of said fifth group,   42 - Apparatus according to claim 40, characterized   characterized in that each position of said fifth group corresponds to a range of a particular characteristic of the cardiac activity data, said data processing device (14) comprising an addressing device   of a particular position of the fifth group in   the detected particular range of the activity signal heart.   43 - Apparatus according to claim 42, characterized   in that said characteristic comprises the interval   between successive peaks of heart beats.   44 - Apparatus according to claim 42, characterized   in that said characteristic comprises the difference   between the intervals associated with cardiac beats c.   Apparatus according to claim 44, characterized   in that the current addressed position of said fourth group (26d) receives the currently determined interval   between beats, calculated while the last * nal of determined interval between beats is cleared.