Electro Cardiogram (ECG) mini reference Haim Shafir Dec 2007 Table of content 1 2
Market analysis ....................................................................................................................... 3 Origins of ECG ....................................................................................................................... 6 2.1 ECG amplification .......................................................................................................... 9 3 Electrical signal propagation in muscle ................................................................................ 10 4 Sinus Node and the Purkinje System of the Heart................................................................ 11 5 Action potential to ECG translation...................................................................................... 12 6 The Einthoven’s triangle....................................................................................................... 13 7 ECG Diagram........................................................................................................................ 15 8 The electrode body interface................................................................................................. 16 8.1 The Electrode ................................................................................................................ 16 8.1.1 Oxidation and Reduction ...................................................................................... 16 8.1.2 Half cell Potential ................................................................................................. 17 8.1.3 Half cell potential table......................................................................................... 17 8.1.4 Perfectly Polarized Electrodes .............................................................................. 18 8.1.5 Perfectly Non-Polarized Electrode ....................................................................... 18 8.1.6 The Silver/Silver Chloride Electrode Ag / AgCl ................................................. 18 8.1.7 Electrode-electrolyte interface and equivalent circuit .......................................... 19 8.1.8 Electrode Skin interface........................................................................................ 20 8.1.9 Simplified circuit .................................................................................................. 20 9 Interference Issues ................................................................................................................ 21 10 Typical ECG sensing circuit ............................................................................................. 23 11 Compression ..................................................................................................................... 24 12 Regulation ......................................................................................................................... 26 12.1 Environmental Testing.................................................................................................. 26 12.2 Software ........................................................................................................................ 27 12.3 Electrical Safety ............................................................................................................ 27 12.4 Electromagnetic Compatibility ..................................................................................... 27 13 Standards & specification. ................................................................................................ 27 13.1 Philips example............................................................................................................. 28 14 Battery Selection............................................................................................................... 32 14.1 Discharge characteristics of lithium batteries ............................................................... 33 15 1.8 volt and Power regulation? ......................................................................................... 34 16 System Design .................................................................................................................. 36 16.1 Block Diagram .............................................................................................................. 36 17 Instrumentation Amplifier Design .................................................................................... 37 17.1 Instrumentation Amplifier References.......................................................................... 38 18 Crystal oscillator ............................................................................................................... 40 18.1 Crystal amplifier ........................................................................................................... 42 1/47
ECG7 primer by HS
3/5/2008
18.2 Crystal References ....................................................................................................... 42 19 Pacemaker Pulse detection................................................................................................ 43 19.1 Pacemaker pulse detection references .......................................................................... 47
2/47
ECG7 primer by HS
3/5/2008
1
Market analysis
From chart 1 and chart 2 one can surmise that there are 5M heart disease live discharges per year. If half have CHD per chart 3, there are 2.5 potential customers for telemetry. chart 4 confirms this number. If 20% of those patients utilize telemetry the total available market is 5000,000 units per year. 20% market share will yield base line sales of 100,000 units per year. In 2004 there were 72,648,000 physician office visits with a primary diagnosis of CVD (NCHS, NHAMCS).and 6,369,000 outpatient department visits with a primary diagnosis of CVD (NHAMCS). If 50% were CHD related and each patient managed 10 visits, the hypothetical number of patients is 4M. If 20% of those need telemetry and the market share for those is 20%, the number of units to this segment is 160,000. The total number of units at 20% market share is 260,000. The market can be expanded to Sleep Apnea detection [3], patient activated event monitoring and pulse oximetry. This number is based on information from an advocacy group [1] and must be taken with a grain of salt. If we add the numbers of afflicted people that each advocacy group claims; each and every citizen of the great county is suffering from multiple afflictions, sick as a dog and just barely manage to escape painful death [2].
Discharges in Millions
7 6 5 4 3 2 1 0 04
00
90
80
70
Years
Chart 1 Hospital discharges for cardiovascular diseases. (USA: 1970–2004) Include those inpatients discharged alive, dead, or status unknown. Source: National Hospital Discharge Survey, NCHS and NHLBI.
3/47
ECG7 primer by HS
3/5/2008
Chart 2 Deaths from diseases of the heart (USA: 1900–2004) Source: Respective NVSR reports. NCHS and NHLBI.
Chart 3 Percentage breakdown of deaths from CVDs (USA: 2004) Source: NCHS and NHLBI
4/47
ECG7 primer by HS
3/5/2008
Discharges in Thousands
1400 1200 1000 800 600 400 200 0 70
80
90
00
04
Years Males
Females
Chart 4 Hospital discharges for coronary heart disease by se, (USA: 1970–2004). Include those inpatients discharged alive, dead, or under unknown status. Source: NHDS, NCHS, and NHLBI
Marketing References [1] Heart Disease and Stroke Statistics—2007 Update: A Report from the American Heart Association Statistics Committee. [2] Bob Garfield, NPR, On the Media, 11/30/2007 [3] Home Diagnosis of Sleep Apnea Chest 2003;124;1543-1579
5/47
ECG7 primer by HS
3/5/2008
2
Origins of ECG
In 1787 Luigi Galvani observed that a frog muscle twitched when exposed to an electrical spark. After Nobili and others developed a sensitive galvanometer in 1825 it was possible to prove that there are charges and currents within the frog itself. DuBois-Reymond introduced the term action potential. In 1887 Augustus Desiree Waller first recorded electric potentials associated with the beating heart from the body surface .Waller carried out his studies using the capillary electrometer invented by the French physicist Gabriel Lippmann in 1873. Willem Einthoven was a Dutch doctor and physiologist. He invented the first practical electrocardiogram (ECG or EKG) in 1903 and received the Nobel Prize in Medicine in 1924 for it. The first electrocardiograph was not user friendly; it weighted about 600 pounds and required five people for operation. The patient did not look too happy either. (see figure 1)
Figure 1
6/47
ECG7 primer by HS
3/5/2008
The measuring device was a string galvanometer (see figure 2), it comprised a silver coated quartz string placed in a field generated by a strong electromagnet. A projection system displayed the sting position on a screen. The shadow moved a millimeter when a current of 10pA flowed through the string.
figure 2
7/47
ECG7 primer by HS
3/5/2008
The first table electrocardiograph manufactured by the Cambridge Scientific Instrument Company of London in 1911 (see figure 3). The electrodes were big buckets of salt water. On the right hand side the arch lamp, in the centre on the table the string galvanometer, and below the switching board for the leads, next left to the camera the timer (rotating wheel with spokes), and on the left hand side the falling-plate camera.
Figure 3
8/47
ECG7 primer by HS
3/5/2008
2.1
ECG amplification
In 1928 Ernestine and Levine reported in the American Heart Journal the use of vacuum tubes to amplify the electrocardiogram instead of the mechanical amplification of the string galvanometer. In the same year Frank Sanborn’s company introduced a portable ECG machine, the weight was 50 pounds and it was powered by a 6 V automobile Battery. One of the first amplifier type electrocardiographs in Europe was developed by Siemens and Halske in Germany in 1934 (see Figure 4).
Figure 4
9/47
ECG7 primer by HS
3/5/2008
3
Electrical signal propagation in muscle
A piece of a living ventricular muscle is placed in a salt solution. Electrodes are on either side of the muscle, no potential difference would be recorded between the two electrodes when the muscle is in its polarized, resting state (upper panel). The outside of the cells is positive relative to the inside, the resting membrane potential is be about -90 mV; no currents will flow along the surface of the muscle and through the bath. If the left side of the muscle is stimulated electrically to induce self-propagating action potentials, a wave of depolarization would sweep across the muscle from leftto-right (lower panel). Midway through this depolarization process, cells on the left (depolarized cells) would be negative on the outside relative to the inside, while non-depolarized cells on the right of the muscle would still be polarized (positive on the outside). A potential difference between the positive and negative electrodes now exists. By convention, a wave of depolarization heading toward the positive electrode is recorded as a positive voltage (upward deflection in the recording). After the wave of depolarization sweeps across the entire muscle mass, all the cells on the outside are negative, and once again, no potential difference would exist between the two electrodes.
Flow of Current in the Chest from the Partially Depolarized Activated Ventricles
10/47
ECG7 primer by HS
3/5/2008
4
Sinus Node and the Purkinje System of the Heart
For a nice animation of the propagation of the signal and its relation to ECG please go to http://upload.wikimedia.org/wikipedia/commons/e/e5/ECG_principle_slow.gif or download ecgsim
11/47
http://www.ecgsim.org/
ECG7 primer by HS
3/5/2008
5
Action potential to ECG translation
12/47
ECG7 primer by HS
3/5/2008
6
The Einthoven’s triangle
The stretch between two limb (arm or leg) electrodes is called a lead. Einthoven named the leads between the three limb electrodes "standard lead I, II and III" referring to the two arm electrodes and the left leg electrode. He studied the relationship between these electrodes, forming a triangle where the heart electrically constitutes the null point. The relationship between the standard leads is called Einthoven's triangle.
13/47
ECG7 primer by HS
3/5/2008
V1: V2: V3: V4: V5: V6:
14/47
right 4th intercostal space left 4th intercostal space halfway between V2 and V4 left 5th intercostal space, mid-clavicular line horizontal to V4, anterior axillary line horizontal to V5, mid-axillary line
ECG7 primer by HS
3/5/2008
7
ECG Diagram
Normal ECG
Sudden cardiac death, monitor recording.
15/47
ECG7 primer by HS
3/5/2008
8
The electrode body interface
8.1
The Electrode
8.1.1
Oxidation and Reduction
The term, oxidation , was derived from the observation that almost all elements reacted with oxygen to form compounds called, oxides. A typical example is the corrosion of iron as described by the chemical equation: 4 Fe + 3 O2 2 Fe2O3 Reduction, was the term originally used to describe the removal of oxygen from metal ores, which "reduced" the metal ore to pure metal: 2 Fe2O3 + 3 C 3 CO2 + 4 Fe Based on the two examples above, oxidation can be defined very simply as, the "addition" of oxygen; and reduction, as the "removal" of oxygen. The original view of oxidation and reduction is that of adding or removing oxygen. An alternative view is to describe oxidation as the losing of electrons and reduction as the gaining of electrons. This electron view of oxidation and reduction helps you deal with the fact that "oxidation" can occur even when there is no oxygen! The definition of redox reactions is extended to include other reactions with nonmetals such as chlorine and bromine. For example, the reaction Mg + Cl2 Mg2+ + 2ClMagnesium loses electrons and is therefore said to be "oxidized", whereas the chlorines gain electrons and are said to be reduced. Another way to judge that the chlorine has been reduced is the fact that the charge on the atoms is made more negative, or reduced. Oxidation is dominant when the current flow is from electrode to electrolyte, and reduction dominate when the current flow is in the opposite Oxidation
Reduction
Current flow
Current flow
16/47
ECG7 primer by HS
3/5/2008
8.1.2
Half cell Potential
Half-Cell potential is determined by the metal involved, Concentration of its ion in solution and Temperature
8.1.3
Half cell potential table
Li+(aq) + e- g Be2+(aq) + 2 eZn2+(aq) + 2 eFe2+(aq) + 2 eSn2+(aq) + 2 ePb2+(aq) + 2 e2 H+(aq) + 2 eCu2+(aq) + eAgCl(s) + eCu2+(aq) + 2 eFe3+(aq) + eAg+(aq) + eAu3+(aq) + 3 eF2(g) + 2 e-
Li(s) Be(s) Zn(s) Fe(s) Sn(s) Pb(s) H2(g) Cu+(aq) Ag(s) + Cl-(aq) Cu(s) Fe2+(aq) Ag(s) Au(s) F-(aq)
-3.05 -1.85 -0.76 -0.44 -0.14 -0.13 0.00 +0.13 +0.22 +0.34 +0.77 +0.80 +1.50 +2.87
Standard Hydrogen electrocde ECG electrode
See Nernst equation
17/47
ECG7 primer by HS
3/5/2008
8.1.4
Perfectly Polarized Electrodes
Electrodes in which no actual charge crosses the electrode-electrolyte interface when a current is applied. The current across the interface is a displacement current and the electrode behaves like a capacitor. Example : Platinum electrode used in stimulation 8.1.5
Perfectly Non-Polarized Electrode
Electrodes in which current passes freely across the electrode-electrolyte interface, requiring no energy to make the transition. Example: Ag/AgCl Electrode used in monitoring 8.1.6
The Silver/Silver Chloride Electrode Ag / AgCl
Close to perfect non-polarized electrode, it comprises silver metal coated with a thin layer of ionic compound of AgCl, it is stable in liquid that has large quantity of Cl- such as biological fluid. Electrode reaction oxidation of silver atoms to silver ions: Ag Ag+ + ecombination of silver ions with chloride ionic compound AgCl:
Ag+ + Cl-
AgCl
AgCl slightly soluble to water, it will be deposited on the surface of the electrode, will quickly saturate and come to equilibrium.
18/47
ECG7 primer by HS
3/5/2008
8.1.7
19/47
Electrode-electrolyte interface and equivalent circuit
ECG7 primer by HS
3/5/2008
8.1.8
Electrode Skin interface
8.1.9
Simplified circuit
Cd Vh*
Rep||Rp
Vh*
ZE
Rd
20/47
ECG7 primer by HS
3/5/2008
9
Interference Issues
Electrode Motion noise, movement of the patient, induces pressure variations at the skin electrode interface which generates low frequency noise in the signal present at the amplifier input. This mechanism has been analyzed by Zipp and Ahrens [1] Very low frequency drift (base line winder) in the polarization potential induced by the presence of sweat and other factors.[10] Electrical Interference is superimposed biological signal. Some can be filtered out as it is out-ofband, but most is often in-band, particularly that caused by the mains power supply. Mains hum can be induced by electromagnetic and electrostatic induction. [2],[3]. The magnetic field associated with the mains supply current flowing in nearby electrical equipment cuts the loop enclosed by the subject, the electrode leads and the amplifier and induces current in the leads which directly proportional to the area of the loop. In a cheats mounted monitor the loop area is small and the interference in not significant. The electric field associated with the mains supply is capacitively coupled to the subject who is also coupled to ground via the body capacitance as shown in the figure below:
21/47
ECG7 primer by HS
3/5/2008
This is salient in battery-operated instruments when the common supply line of the amplifier is not at true earth potential and an isolation capacitance is present [4]. A displacement current flows through the subject to ground, splitting between the two paths as shown. This current generates an interfering signal at each of the electrodes relative to ground. When the electrodes are mounted close together, the differential component of the interfering signal is minimal and it is mostly common-mode. The body and isolation capacitances have similar magnitudes [4],[5] and around 50Hz they have a reactance of about 15 M , which is greater than the body and electrode impedances. The common-mode potential at either electrode with respect to the amplifier common can be estimated as: VCM = (ID/2)*ZE. Displacement currents of the order of 0.5 μA have been measured [2],[6],[7] This gives a common-mode interfering signal level of 37.5 mV. If the minimum input ECG signal level to the amplifier is 100μV and the error at the output due to common mode signal is to be less than 5% a CMRR of 77dB is required. 20*LOG(37.5m/5u) Van Rijn [8] provides a detailed coupling model and in depth discussion of interference. In Email communication R. Yazicioglu claims that AC coupling and very high CMRR are sufficient and DRL (Driven Right Leg) circuit is not needed. [9] [1] P. Zipp and H. Ahrens, “A model of bioelectrode motion artifact and reduction of artifact by amplifier input stage design,” J. Biomed. Eng., vol. 1, pp. 273–276, 1979 [2] J. C. Huhta and J. G. Webster, “60-Hz interference in electrocardiography,” IEEE Trans. Biomed. Eng., vol. BME-20, pp. 91–101, Mar. 1973. [3] B. B. Winter and J. G. Webster, “Reduction of interference due to common mode voltage in biopotential amplifiers,” IEEE Trans. Biomed. Eng., vol. BME-30, pp. 58– 62, Jan. 1983 [4] B. B. Winter and J. G. Webster “Driven-right-leg circuit design,” IEEE Trans. Biomed. Eng., vol. BME-30, pp. 62–66, Jan. 1983. [5] N. V. Thakor and J. G. Webster, “Ground-free ECG recording with two electrodes,” IEEE Trans. Biomed. Eng., vol. BME-27, pp. 699–704, Dec. 1980. [6] Martin J. Burke and Denis T. Gleeson “A Micropower Dry-Electrode ECG Preamplifier” IEEE Trans. Biomed. Eng, Vol. 47, No. 2, Feb 2000 [7] A. C. Metting van Rijn, A. Peper, and C. A. Grimbergen, “High-quality recording of bioelectric events. Part I: Interference reduction, theory and practice,” Med. Bio. Eng. Comput., vol. 28, pp. 389–397, Sep. 1990. [8] R firat Yazicioglu, p merken “a 60 μw 60 nv/√hz readout front-end” ieee jssc 2007 vol. 42, no. 5, may 2007 [9] R Warlar, C Eswaran – Filter for base line wander Medical and Biological Engineering and Computing, 1991 - Springer
22/47
ECG7 primer by HS
3/5/2008
10 Typical ECG sensing circuit
Typical solutions today uses an instrumentation amplifier such as the INA118, a common mode feedback drives the shield and the Leg terminal. Usually a band pass and a 50Hz notch filters are used before sampling the signal with an A/D. The 50Hz notch may be digital. The European open ECG standard (http://www.openecg.net) recommends sampling rate of 500 samples per second and a LSB of 5μV
23/47
ECG7 primer by HS
3/5/2008
11 Compression
Lossless compression can provide a Compression Ratio (CR) of 2-3, see table 1, and reference [1]. Lossy compression provides higher CR with a percent residual difference (PRD) of 3% to 25%; see table 2 reference [2]. Lossless compression with a CR of 2-3 will not enable a single chip implementation without external memory. Lossy compression is more effective but risky. Use of compression needs to be proven fail safe; any diagnostic error due to compression error is a liability. The European open ECG forum [3] tried to specify a compression method but it seems there are no clear suggestions for compression.
Table 1
24/47
ECG7 primer by HS
3/5/2008
Table 2
[1] ECG Signal Compression Based on Burrows-Wheeler Transformation and Inversion Z Arnavut -. IEEE Tran on Biomed Eng Vol. 54, No. 3, March 2007 [2] Mean-shape vector quantizer for ECG signal compression Cardenas-Barrera, J.L. Lorenzo-Ginori, J.V. IEEE Tran on Biomed Eng, Jan 1999 Vol: 46, Issue: 1 [3] How To Implement SCP-ECG Part II Hannover, July 2003 http://www.openecg.net/
25/47
ECG7 primer by HS
3/5/2008
12 Regulation An ECG monitor is a class 2 medical device regulated by the U.S. Department Of Health And Human Services, Food and Drug Administration, Center for Devices and Radiological Health. The device is subject to Section 510(k) of the Food, Drug and Cosmetic Act; it requires the device manufacturers to notify FDA, at least 90 days in advance, of their intent to market a medical device. This is known as Premarket Notification - also called PMN or 510(k). It allows FDA to determine whether the device is equivalent to a device already placed into one of the three classification categories. The Federal Food, Drug, and Cosmetic Act (the act), as amended by the Medical Device Amendments of 1976 (the 1976 amendments) (Public Law 94-295) and the Safe Medical Devices Act of 1990 (the SMDA) (Public Law 101-629), and the Food and Drug Administration Modernization Act of 1997 (FDAMA) (Public Law 105-115). The act, as amended by the 1976 amendments, the SMDA and the FDAMA, establishes a comprehensive system for the regulation of medical devices intended for human use. Section 513 of the act (21 U.S.C. 360c) established three categories (classes) of devices, depending on the regulatory controls needed to provide reasonable assurance of their safety and effectiveness. The three categories of devices are class I (general controls), class II (special controls), and class III (premarket approval). Devices that were not in commercial distribution prior to May 28, 1976, generally referred to as postamendment devices, are classified automatically by statute (section 513(f) of the act) into class III without any FDA rulemaking process. Those devices remain in class III and require premarket approval, unless and until the device is reclassified into class I or II or FDA issues an order finding the device to be substantially equivalent, in accordance with section 513(i) of the act, to a predicate device that does not require premarket approval. The agency determines whether new devices are substantially equivalent to previously offered devices by means of premarket notification procedures in section 510(k) of the act (21 U.S.C. 360(k)) and 21 CFR part 807 of the regulations. 12.1 Environmental Testing
The manufacturer should evaluate the ability of the device to function after exposure to the environmental hazards expected when used by an abusive user. Tests for some of these hazards may be found in EC11, IEC 601-1, UL 2601, IEC 68-2, and IEC 529. If a device is intended to be used outside of the hospital environment (e.g., those indicated for use in a transport environment such as an ambulance or helicopter), it may require additional testing. For example, devices intended for use in an ambulance should generally meet an appropriate shock/vibration test (e.g., see the Diagnostic ECG Guidance, Version 1.0 (November 2, 1998) Page 6 IEC 68-2 series), and should demonstrate immunity to a field strength of 20 V/m (rather than the 3 V/m typically needed
26/47
ECG7 primer by HS
3/5/2008
12.2 Software
Depending on the proposed indications for use, cardiac monitors may be considered to have a level of concern ranging from minor to moderate. Refer to the “Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices” for additional information about software documentation for a 510(k). 12.3 Electrical Safety
Any appropriate standard for electrical safety (e.g., ANSI/AAMI ES-1, IEC 601-1) may be used. If the EC11 standard is used, the manufacturer should conform to the standard’s requirements or justify any deviation from the standard. 12.4 Electromagnetic Compatibility
Electromagnetic compatibility (EMC) testing should be done to demonstrate that the device will not adversely interfere with the performance of other electronic devices (emissions), and will perform as expected in the presence of other electronic devices or other sources of electromagnetic interference (EMI) in the intended environment of use (immunity). NOTE: If the device is intended for use outside the hospital environment, additional testing may be necessary 13 Standards & specification. 1.
ANSI/AAMI EC11-1991, “Diagnostic Electrocardiographic Devices”
2.
ANSI/AAMI EC12:2000, Disposable ECG Electrodes
3.
ANSI/AAMI EC13-1992, “Cardiac monitors, heart rate meters, and alarms”
4.
ANSI/AAMI EC38-1994, “Ambulatory Electrocardiographs”
5.
ANSI/AAMI ES1-1993, “Safe current limits for electromedical apparatus
6.
IEC 60601-1-2 (1993), “Medical Electrical Equipment- Part 1: General Requirements for Safety; 2. Collateral Standard: Electromagnetic Compatibility- Requirements and Tests”.
7.
IEC 68-2-6 1982) with Amendment 1-1983 and Amendment 2-1985: Basic Environmental Testing Procedures part 2: a. Test Fc and Guidance: Vibration b. Test Ea and Guidance: Shock c. Test Fdc and Guidance: Random Vibration Wide Band
The European standards also need consideration. For example The IEC 60601-1-2 has been adopted in Europe as the European Norm (EN 60601-1-2). the next few pages show examples from Philips DigiTrak
27/47
ECG7 primer by HS
3/5/2008
13.1 Philips example
Table A-1. Guidance and Manufacturer’s Declaration: Electromagnetic Emissions
The DigiTrak Plus recorder is intended for use in the electromagnetic environment specified in the table below. The customer or the user of the DigiTrak Plus recorder should assure that it is used in such an environment.
28/47
Emissions Test
Compliance
Electromagnetic Environment: guidance
RF Emissions CISPR 11
Group 1
The DigiTrak Plus recorder uses RF energy only for its internal function. Therefore, its RF emissions are very low are not likely to cause any interference in nearby electronic equipment.
RF Emissions CISPR 11
Class B
The DigiTrak Plus recorder is suitable for use in all establishments, including domestic establishments and those directly connected to the public low-voltage power supply network that supplies buildings used for domestic purposes.
Harmonic Emissions IEC 61000-3-2
Not Applicable
Voltage fluctuations/ flicker emissions IEC 61000-3-3
Not Applicable
ECG7 primer by HS
3/5/2008
Table A-2. Guidance and Manufacturer’s Declaration: Electromagnetic Immunity
The DigiTrak Plus recorder is intended for use in the electromagnetic environment specified below. The customer or the user of the DigiTrak Plus recorder should assure that it is used in such an environment. Immunity Test
IEC 60601 Test Level
Compliance Level Electromagnetic Environment: Guidance
Electrostatic Discharge (ESD) IEC 61000-4-2
+/- 6 kV contact +/- 8 kV air
Complies
Floors should be wood, concrete, or ceramic tile. If floors are covered with synthetic material, the relative humidity should be at least 30%.
Electrical Fast transient/burst IEC 61000-4-4
+/- 2 kV for power supply line +/- 1 kV for input/ output lines
N/A
The DigiTrak Plus does not have AC or DC power lines.
Surge IEC 610004-5
+/- 1 kV differential mode +/- 2 kV common mode
N/A
The DigiTrak Plus does not have AC or DC power lines
short interruptions and voltage variations on power supply input lines IEC 61000-4-11
<5% UT (>95% dip in UT) for 0.5 cycle 40% UT (60% dip in UT) for 5 cycles 70% UT (>30% dip in UT) for 25 cycles <5% UT (>95% dip in UT) for 5 seconds
N/A
The DigiTrak Plus does not have AC or DC power lines
Complies
Power frequency magnetic fields should be at levels characteristic of a typical location in a typical commercial or hospital environment
Power frequency (50./60 Hz) magnetic field IEC 61000-4-8
3 A/m
NOTE UT is the AC mains voltage prior to application of the test level.
29/47
ECG7 primer by HS
3/5/2008
Table A-3. Guidance and Manufacturer’s Declaration: Electromagnetic Immunity
The DigiTrak Plus recorder is intended for use in the electromagnetic environment specified below. The customer or the user of the DigiTrak Plus recorder should assure that it is used in such an environment. Immunity Test
IEC 60601 Test Level
Compliance Level
Electromagnetic Environment: guidance
Conducted RF IEC 61000-4-6 Radiated RF IEC 61000-4-3
3 Vrms 150 kHz to 80 MHz 3 V/m 80 MHz to 2,5 GHz
3 Vrms 3 V/m
Portable and mobile RF communications equipment should be used no closer to any part of the DigiTrak Plus, including cables, than the recommended separation distance calculated from the equation applicable to the frequency of the transmitter. Recommended separation distance D = 1.2vP D = 1.2vP 80 MHz to 800 MHz D = 2.3vP 800 MHz to 2.5 GHz Where P is the maximum output power rating of the transmitter in watts (W) according to the transmitter manufacturer and D is the recommended separation distance in metres (m). Field strengths from fixed RF transmitters, as determined by an electromagnetic site survey, should be less than the compliance level in each frequency range. Interference may occur in the vicinity of equipment marked with the following symbol:
NOTE 1 At 80 MHz and 800 MHz, the higher frequency range applies. NOTE 2 These guidelines may not apply in all situations. Electromagnetic propagation is affected by absorption and reflection from surfaces, objects, and people. Additional notes are on following page.
30/47
ECG7 primer by HS
3/5/2008
Table A-4. Recommended Separation Distances Between Portable and Mobile RF Communications Equipment and the DigiTrak Plus recorder: for equipment and systems that are not life-supporting
The DigiTrak Plus recorder is intended for use in the electromagnetic environment in which radiated RF disturbances are controlled. The customer or the user of the DigiTrak Plus recorder can help to prevent electromagnetic interference by maintaining a minimum distance between portable and mobile RF communications equipment (transmitters) and the DigiTrak Plus recorder as recommended below, according to the maximum output power of the communications equipment. Rated Maximum Output Power of Transmitter W
Separation Distance According to Frequency of Transmitter (m)
150 KHz to 80 MHz D = 1.2vP
80 MHz to 800 MHz D = 1.2vP
800 MHz to 2.5 GHz 2.3vP
0.01
.12
.12
.23
0.1
.38
.38
.73
1
1.2
1.2
2.3
10
3.8
3.8
7.3
100
12.0
12.0
23.0
For transmitters rated at a maximum output power not listed above, the recommended separation distance D in meters (m) can be estimated using the equation applicable to the frequency of the transmitter, where P is the maximum output power rating of the transmitter in watts (W) according to the transmitter manufacturer. NOTE 1 At 80 MHz and 800 MHz, the separation distance for the higher frequency range applies. NOTE 2 These guidelines may not apply in all situations. Electromagnetic propagation is affected by the absorption and reflection from structures, objects, and people.
31/47
ECG7 primer by HS
3/5/2008
14 Battery Selection Most coin type lithium batteries are available in two types: poly carbon monofluoride lithium batteries (BR series) for uses that require extended reliability and safety, and manganese dioxide lithium batteries (CR series) for uses that require high voltage and strong load pulse characteristics. BR type is usually selected for medical applications. Table 1 - (CF)n/LI: Poly-Carbon Monofluoride (BR)
Model
Nominal Voltage (V)
Nominal Capacity (mAh)
Continuous Drain (mA)
Diameter (mm)
Height (mm)
Weight (g)
BR1220
3
35
0.03
12.5
2.00
0.7
BR1225
3
48
0.03
12.5
2.50
0.8
BR1632
3
120
0.03
16.0
3.20
1.5
BR2032
3
190
0.03
20.0
3.20
2.5
BR2325
3
165
0.03
23.0
2.50
3.2
BR2330
3
255
0.03
23.0
3.00
3.2
BR3032
3
500
0.03
30.0
3.20
5.5
Table 2 - Mn02/LI:Manganese Dioxide (CR)
Nominal Voltage (V)
*Nominal Capacity (mAh)
Continuous Drain (mA)
Diameter (mm)
Height (mm)
Weight (g)
CR1220
3
35
0.10
12.5
2.00
1.2
CR1616
3
55
0.10
16.0
1.60
1.2
CR1632
3
140
0.10
16.0
3.20
1.8
CR2032
3
225
0.20
20.0
3.20
2.9
CR3032
3
500
0.20
30.0
3.20
6.8
Model
32/47
ECG7 primer by HS
3/5/2008
14.1 Discharge characteristics of lithium batteries
The BR-type lithium chemistry uses poly carbon monofluoride (CF)n as the positive material, which results in a stable voltage throughout the life of the cell and stable performance at comparatively high environmental temperatures. The self-discharge of the BR chemistry at elevated temperatures is also superior to that of other lithium technologies. The dominant applications for BR lithium batteries are real time clock, memory back-up and medical devices. CR lithium batteries produce a tapered discharge profile and perform well in comparatively large current applications. As a result, CR type batteries are preferred for higher current drain, intermittent pulse mode applications such as alarm actuation or remote keyless entry systems. CR lithium technology also offers a small cost advantage over BR type batteries. It is also slightly heavier than BR type batteries
33/47
ECG7 primer by HS
3/5/2008
15 1.8 volt and Power regulation? If a 1.8 volt design is needed, a micro power step down regulator can be used to provide 1.8 volt from a 3v lithium cell. A capacitor only DC to DC step down regulator will be about 70% efficient and noisy. Such a device is LTC 1503-1.8 Low Noise, High Efficiency, and Inductor less Step-Down DC/DC Converter. http://www.linear.com
A more efficient choice is a buck step down regulator; it will need a 2.2μH inductor and can achieve efficiencies of 85%. Such a device is TI TPS62261, 1.8v, 2.25 MHz 600 mA Step Down Converter in 2x2SON/TSOT-23 Package. http://focus.ti.com/lit/ds/symlink/tps62261.pdf
34/47
ECG7 primer by HS
3/5/2008
Taiyo Yuden CKP25202R2M-T is a 2.2μH 2.5x2x1mm flat inductor http://www.yuden.co.jp/ut/product/inductor/CKP25202R2M-T.html
Wurth 744030002 is 3.3x3.3x1mm 2.2μH inductor.
http://www.wuerth-elektronik.de
It is not clear at his point if the effort to work in 1.8 volt is justified. The regulator consumes power, the switching noise could alias into the sampled signal, EMI could be an issue. The input amplifier and the A2D will be easier to design and will probably have better performance in 3volt.
35/47
ECG7 primer by HS
3/5/2008
16 System Design 16.1 Block Diagram
electrodes
IA
Comm. Cntl
12bit A/D 500 S/s
RTC
Power Monitor XC
Control CPU?
Interface
lithium battery
PLL
A single channel monitor generally comprise the following circuits, any custom chip will need to provide more than a single channel. • • • • • • • • •
36/47
Instrumentation amplifier, high gain, high input impedance, high CMRR differential amplifier. 12 bit, 500 samples/sec , 5μV LSB low power analog to digital converter, this is not sufficient for pacemaker pulse detection or signal-averaged ECG. Communication Controller Power monitor, monitors the battery voltage and indicates battery empty event RTC, Real time clock XC, Low frequency low power Crystal oscillator PLL, optional PLL to generate the memory controller clock if using a low frequency xtal Interface, provides programming, memory access, indicators and control. Control, can be a logic or CPU, sequences sampling and recording, provides power management.
ECG7 primer by HS
3/5/2008
17 Instrumentation Amplifier Design High-quality representation of Q- and R-waves is possible if the bandwidth is 0.05 Hz to 150 Hz, input range should be at least ±5mV; the least significant bit, no more than 5 μV; sampling frequency, at least 500 per sec; time shift between channels (leads), no more than 100 msec. The requirements for signal-averaged ECG systems are pass band of 0.05 to300 Hz; LSB of 2.5 μV (desirably, 0.3 μV); sampling frequency at least 1000 samples/sec; time shift between channels no more than 50 msec (desirably, zero) and resolution of at least 12 bits (desirably, 14-16). [25] Pacemaker pulse detection requires higher bandwidth and sample rates of 2000 to 10000 samples per second for DSP based detection.
Frequency and amplitude characteristics of biopotential signals [2] The state of the art design is a Chopper Stabilized Instrumentation Amplifier. The Chopping significantly improves the 1/f noise and input offset [3]. T Denison [1] and Refet Firat Yazicioglu [2] published the latest papers and will also present new papers in ISSCC 2008, SF. Denison in [6] provided this table.
Parameter / Paper Supply Current Consumption Supply (V) NEF Gain @ 10Hz Input Referred Noise Noise Band Width CMRR (50/60Hz) High Pass –3dB frequency Low Pass –3dB frequency
37/47
[1] 1.2 1.8 4.9 45 0.93 100 105 0.5 250
ECG7 primer by HS
[2] 60 +/-1.5 9.2 80 0.86 150 117 0.3 150
[4] 20 3V 7.8 60 0.45 40 110 0.5 150
unit uA V dB uVrms Hz dB Hz Hz
3/5/2008
17.1 Instrumentation Amplifier References
[1]
Timothy Denison, Kelly Consoer “A 2.2μW 94nV/√Hz chopper-stabilized instrumentation amplifier for chronic measurement of bio-potentials” IMTC 2007
[2]
Refet Firat Yazicioglu, Patrick Merken “A 60 _W60 nV/ Hz Readout Front-End for Portable Biopotential Acquisition Systems” IEEE JSSC Vol 42, No 5, May 2007
[3]
CC Enz, GC Temes “Circuit Techniques for Reducing the Effects of Op-Amp Imperfections: Autozeroing, Correlated Double Sampling, and Chopper Stabilization” Proceedings of the IEEE, Nov 1996
[4]
K.A. Ng and P.K. Chan, "A CMOS Analog Front-End IC for Portable EEG/ECG Monitoring Applications," IEEE Trans. On Circuits and Systems, vol. 52, no. 11, 2005.
[5]
Bakker, K. Thiele, and J. H. Huijsing, “A CMOS nested-chopper instrumentation amplifier with 100-nV offset,” IEEE J. Solid-State Circuits, vol. 35, no. 12, Dec. 2000
[6]
Timothy Denison, Kelly Consoer “A 2.2μW 94nV/√Hz, Chopper-Stabilized Instrumentation Amplifier for EEG Detection in Chronic Implants IEEEE ISSCC Feb 2007
[7]
M. S. J. Steyaert, W. M. C. Sansen, and C. Zhongyuan, “A micropower low-noise monolithic instrumentation amplifier for medical purposes,” IEEE J. Solid-State Circuits, vol. sc-22, no. 6, pp. 1163–1168, Dec. 1987
[8]
Donald A. Kerth, Douglas S. Piasecki “An Oversampling Converter for Strain Gauge Transducers” IEEE JSSC Vol 21, No 12, Dec 1992
[9]
Andrea Gerosa, “A Fully Integrated Two-Channel A/D Interface for the Acquisition of Cardiac Signals in Implantable Pacemakers” IEEE JSSC, Vol. 39, No. 7, July 2004
[10]
M. Zimmermann “A CMOS-based Sensor Array System for Chemical and Biochemical Applications” Proceedings of ESSCIRC, Grenoble, France, 2005
[11]
A Bilotti, G Monreal “Chopper-Stabilized Amplifiers with a Track and Hold” IEEE Trans on Circuits and Systems, Vol 46, No 4, April 1999
[12]
Chan, Ng “A CMOS Chopper-Stabilized Differential Difference Amplifier for Biomedical Integrated Circuits” The 47th IEEE International Midwest Symposium on Circuits and Systems
[13]
Wu, Zu “ Low-Voltage Low-Noise CMOS Instrumentation Amplifier for Portable Medical Monitoring Systems” IEEE 2005
[14]
Burt, Zhang “A Micropower Chopper-Stabilized Operational Amplifier Using a SC Notch Filter With Synchronous Integration Inside the Continuous-Time Signal Path” IEEE JSSC Vol 41, No 12, Dec 2006
[15]
J. H. Nielsen, E Bruun “A CMOS Low-Noise Instrumentation Amplifier Using Chopper Modulation” 2004 Springer Science
[16]
E. L. Douglas, D.F. Lovely, “A Low-Voltage Current-Mode Instrumentation Amplifier Designed in a 0.18-Micron CMOS” CCGEI 2004
38/47
ECG7 primer by HS
3/5/2008
[17]
A. Uranga1, N. Lago “A LOW NOISE CMOS AMPLIFIER FOR ENG SIGNALS” ISCAS 2004
[18]
Chih-Jen Yen, Wen-Yaw Chung “Micro-Power Low-Offset Instrumentation Amplifier IC Design for Biomedical System Applications”Ieee transactions on circuits and systems, vol. 51, no. 4, april 2004
[19]
Enrique Mario Spinelli “AC-Coupled Front-End for Biopotential Measurements” ieee transactions on biomedical engineering, vol. 50, no. 3, march 2003
[20]
Andrew T. K. Tang “A 3μV-Offset Operational Amplifier with 20nV/√Hz Input Noise PSD at DC Employing both Chopping and Autozeroing” 2002 IEEE International SolidState Circuits Conference.
[21]
Gary m. Friesen, t c. Jannett “a comparison of the noise sensitivity of nine qrs detection algorithms” ieee trans on biomedical eng. vol. 37. no. i. january 1990
[22]
AC Metting van Rijn , “high-quality recording of bioelectric events, interference reduction, theory and pratice Part 1”, medical & biological eng & computing, 1990
[23]
AC Metting van Rijn , “high-quality recording of bioelectric events theory and pratice Part 2”, medical & biological eng & computing, 1991.
[24]
Bernhard Fuchs, Sven Vogel “Universal application-specific integrated circuit for bioelectric data acquisition” Medical Engineering & Physics 24 (2002)
[25]
IEC 60601-2-51. Medical Electrical Equipment. Part 2- 51: Particular Requirement for Safety, Including Essential Performance, of Recording and Analyzing Single Channel and Multichannel Electrocardiographs, New York (2003)
39/47
ECG7 primer by HS
3/5/2008
18 Crystal oscillator The clock frequencies of most digital system today are in the MHz range. A basic ECG monitor needs to sample the physiological signal at 500Hz. If chopping and over sampling sigma delta analog to digital converter are used, frequencies in the order of tens of KHz are needed. To provide the lowest power consumption and the smallest use of printed circuit board real estate a tuning fork crystal should be used. An advanced device with higher sample rate may need to use high frequency crystal which is larger in size.
Table 1 shows the difference between an AT and XY cut.
Cut
vibration mode
Frequency range
Shunt to notional capacitance ratio
AT
Fundamental
2 – 80 MHz
200
XY
Flexural, Tuning fork
16 – 150 KHz
425 - 800
image
Table 1 Citizen CM315 http://www.citizen.co.jp and Fox FX135 http://www.foxonline.com/watchxtals.htm are examples for a tuning fork crystal, they measure 3x1.5x0.9 mm.
40/47
ECG7 primer by HS
3/5/2008
The most popular tuning fork crystal is used for time keeping; it oscillates at a frequency of 32.718 KHz. All chopping, over sampling and other low power digital activities should be at that frequency or its derivatives. A over sampling sigma delta can over sample by 64 to get 500 samples per second (500x64=32K).
CM315 specification .
41/47
ECG7 primer by HS
3/5/2008
18.1 Crystal amplifier
Most implementations use a pierce type oscillator, care should be taken not to over drive the crystal, 1uW is typical for tuning forks. Operation in weak inversion can enable low power operation with reported amplifier current of 1μA [1]. A biased inverted is the starting point, the notional model parameters (see figure on the right) can be extracted from the data sheet and simulation.
18.2 Crystal References
[1]
Eric Vittoz “High-performance crystal oscillator circuits” IEEE JSSCC 1988
[2]
Max Forrer “survey of circuitry for wrist watches” 1972
[3]
J.-M. Friedta “Introduction to the quartz tuning fork” Am. J. Phys. May 2007
[4]
Hiroshi Yoda “Low power crystal oscillators” 1972
42/47
ECG7 primer by HS
3/5/2008
19 Pacemaker Pulse detection Accurate detection of pacemaker pulse locations is required for pulse removal prior to QRS detection, for classification of the paced status of each beat, and for identification of proper pacemaker function or malfunction. The origin of modern cardiac pacing started when the first pacemaker, developed by Dr. Rune Elmqvist, was used in a patient in 1958 by Dr. Ake Senning. In 1959, the engineer Wilson Greatbatch and the cardiologist W.M. Chardack developed the first fully implantable pacemaker. This device was essentially used to treat patients with complete AV block caused by StokesAdams diseases, delivering a single-chamber ventricular pacing. It measured 6 cm in diameter and 1.5-cm thick, and the total weight of the pacemaker was approximately 180 g. The pacemaker circuit delivered 1-ms wide pulses to the electrode, a pulse amplitude of 10 mA and a repetition rate of 60 bpm. [1]
Schematic of first implantable pace maker.
43/47
ECG7 primer by HS
3/5/2008
The first pacemakers provided a high energy pulse with larger amplitude the QRS complex. The pacemaker pulse was 80mv to the 1-2 mv of the complex. At that time ECG devices filtered out the pulse using slew limiting circuits [2] as illustrated in next figures.
than QRS the
Modren, minimal energey pacemakers generate a lower amplitude pulse as demonstrated in the following figure.
Typical example of minimal energy dual-chamber pacemaker pulses in diagnostic bandwidth, 500 samples/sec ECG. Pulse locations are marked with +.
44/47
ECG7 primer by HS
3/5/2008
Traditionally pacemaker pulse detection was done in the front-end of electrocardiograph devices on high bandwidth signals. Many devices use special analog circuitry designed to detect the high signal slew rates typical of pacemaker pulses. These circuits, however, are prone to false detections. (following figure) [5]
Examples of noise spikes detected as pacemaker pulses (+)in clinically acquired ECGs.
Pediatric ECG also presets a challenge due to narrow QRS complex. (Figure on right: ECG of 6 old girl)
month
Other devices use a DSP based software algorithms, they require high bandwidth, high sample rate data, from 2,000 to 10,000 samples per second. According to [6], it is necessary to detect pacemaker pulses with amplitude exceeding 2 mV, front duration less than 100 msec, and total duration exceeding 0.5 msec. The actual duration of a pacemaker pulse reaches 2 msec.
45/47
ECG7 primer by HS
3/5/2008
Typical spike forms in a pacemaker ECG. R represents the location of the R wave, Sp represents the location of the atrial pacing spike, and Sr represents the location of the ventricular pacing spike.
Diagram of pacemaker pulse separation: 1) pass band filter; 2) absolute value unit; 3) window filter; 4) amplitude detector.
High pass output on lower trace.
46/47
ECG7 primer by HS
3/5/2008
19.1 Pacemaker pulse detection references
[1]
Sandro a.p. Haddad “The evolution of pacemakers” IEEE eng in medicine and biology magazine. may/june 2006
[2]
By James L. Larsen HP Journal, April 1972
[3]
D. A. Prilutskii “Broadband Analog-to-Digital Conversion of Electrocardiograms” Biomedical Engineering, Vol. 38, No. 3, 2004, pp. 140-146
[4]
Jing Bai, “A pacemaker working status telemonitoring algorithm” IEEE trasn on information technology in biomedicine, vol. 3, no. 3, september 1999
[5]
Eric D. Helfenbein “A Software-based Pacemaker Pulse Detection and Paced Rhythm Classification Algorithm”, Journal of Electrocardiology Vol. 35 Supplement 2002
[6]
IEC 60601-2-51. Medical Electrical Equipment. Part 2- 51: Particular Requirement for Safety, Including Essential Performance, of Recording and Analyzing Single Channel and Multichannel Electrocardiographs, New York (2003)
47/47
ECG7 primer by HS
3/5/2008
