Journal of Electrocardiology
Volume 42, Issue 6 , Pages 574-579, November 2009

Meet the challenge of high-pass filter and ST-segment requirements with a DC-coupled digital electrocardiogram amplifier

SCHILLER AG, Biomed. Research and Signal Processing, Baar, Switzerland

Received 3 April 2009 published online 24 August 2009.

Article Outline

Abstract 

Background

The high-pass filter (HPF) in an electrocardiogram (ECG) amplifier can distort the ST segment required for ischemia interpretation. Therefore, the current standards and guidelines require −3 dB for monitoring and −0.9 dB for diagnostic purposes at 0.67 Hz. In addition, a minimal reaction to a rectangular pulse of 300 μV has to be proven. We raise the question of why the design of a DC-coupled digital ECG amplifier is reasonable when today the AC-coupled digital ECG amplifier including a 0.05-Hz HPF works so well, meets all required standards, and is already safe. We make the hypothesis that a digital DC-coupled ECG amplifier can as well meet the requirements and guarantee the same safety levels at the same time provide a higher degree of freedom for future improvements of the ECG signal quality.

Methods

Firstly, a historical research of the origin of the 0.05-Hz requirement has been made. Secondly, triangular pulses simulating unipolar QRS complexes have been passed through a digital filter to get qualitative results of the HPF response. And finally, to quantitatively describe the filter response, corresponding test requirement signals have been passed through a digital filter to simulate the HPF behavior, therefore understanding the reasons for the required tests.

Results

The oldest reference found to the 0.05-Hz filter dates from 1937. At that time, DC-coupled analogue ECG amplifiers were used. The simulation of the AC-coupled ECG amplifier with a first-order analogue HPF shows that the rectangular 300-μV pulse is a phase requirement and more restrictive than the frequency requirements. The phase requirement in fact corresponds to the requirement of a 0.05-Hz first-order analogue HPF (−3 dB) even if −0.9 dB at 0.67 Hz is required. The DC-coupled ECG amplifier (without an analogue HPF and during online and off-line acquisition) fulfils the phase and frequency requirements, just as the digital AC-coupled ECG amplifier does.

Conclusions

An AC-coupled ECG amplifier based on a first-order analogue HPF must have a maximum cutoff frequency of 0.05 Hz or requires a phase equalizer causing a delay of the acquired ECG. Because the desired delay during online acquisition should be short, the solution is practical but could be improved. Not the frequency cutoff of the HPF but the phase distortion of such a filter should be discussed. The DC-coupled ECG amplifier is as safe as the AC-coupled ECG amplifier; but it provides a higher degree of freedom for future filter designs certainly improving the ECG signal quality, while the safety can be guaranteed. Furthermore, the DC-coupled ECG amplifier allows investigation of the HPF, which is not easily possible when an AC-coupled ECG amplifier including the HPF is to be investigated.

Keywords: Electrocardiogram, Digital amplifier, DC coupling, High-pass filter

 

Back to Article Outline

Introduction 

Why design a DC-coupled digital electrocardiogram (ECG) amplifier when today the AC-coupled digital ECG amplifier including a 0.05-Hz high-pass filter (HPF) works so well and meets all required standards?1, 2, 3, 4, 5, 6 Moreover, these regulations, as we will show later, have not changed much since 1967. We can therefore conclude that an ECG device available on the market is safe and does not need to be changed. “Never change a running system” is not a bad strategy, and this conclusion is not wrong. Nevertheless, there are some indications that suggest that a change, although it might not be necessary, could improve today's ECG systems' signal quality to make a step forward. At least 1 ECG manufacturer is already selling a device with a 24-bit analogue to digital converter (ADC); further, a patent exists that is based on a 24-bit ADC (US 2002/0111777). A 24-bit ADC means that an ECG with a maximal offset of ±8 V can be converted at a resolution of 1 μV/bit. Thus, there is no longer a need for AC coupling the ECG (ie, eliminate the ECG offset before the ADC). If a 16-bit ADC is used with a 1-μV/bit resolution, only ±32 mV full range can be converted. This means that the ±300-mV ECG offset has to be reduced by use of an adequate HPF. The filter reduces the DC component of the ECG including the ECG offset that is part of this component. To convert an ECG including the ±300 mV at a resolution of 1 μV/bit, at least 19.2 bits are required.

Technically speaking, a 20-bit ADC is enough to avoid the analogue 0.05-Hz HPF, enabling digital DC coupling. As such ADCs are available, a digital DC-coupled ECG amplifier is technically possible.

So one could ask the question, “why are not all ECG device manufacturers already selling ECG devices including such digital DC-coupled ECG amplifiers?” This review article aims to discuss this question and prove our hypothesis that a digital DC-coupled ECG-amplifier is in fact an improvement compared with an AC-coupled ECG amplifier, while the same safety levels can be guaranteed. The topic, as already shown before, can be reduced to the problem of the 0.05-Hz HPF in an ECG amplifier. We will do this by, firstly, reviewing the history of the guidelines and regulations. Secondly, the problem of the 0.05-Hz HPF and the 3-μV impulse response test will be discussed. And finally, we will present the design of a digital DC-coupled ECG amplifier and summarize its advantages over an AC-coupled ECG amplifier.

Back to Article Outline

History research 

The starting point of the history research was the American Heart Association (AHA) ECG recommendation published in 2007/2009 and the current standard list. An additional electronic search has been done to not miss important, recently published information about the performance of the ECG amplifier. Scientific publications have only been taken into account when cited by a standard, guideline, or recommendation, as we make the hypothesis that these committees have already conducted that work before writing the corresponding recommendation, guideline, or standard. From this published information, the cited bibliographies have been taken into account by iteration. Recently published articles have again been taken into account by iteration back to the historical sources of the ECG amplifiers that have been found, forming the end point of the history research.

During the standard research, the following standards have been identified to be relevant for our task: for diagnostic ECG devices, the international standard 60601-2-511 by International Electrotechnical Commission and the EC11 standard4 by the Association for the Advancement of Medical Instrumentation/American National Standards Institute; for monitoring and ambulatory ECG devices, the international standard 60601-2-27 and 60601-2-472, 3 by International Electrotechnical Commission and the EC13 and EC385, 6 standard by the Association for the Advancement of Medical Instrumentation/American National Standards Institute. These standards require −3 dB at 0.67 Hz for monitoring ECG devices and −0.9 dB at the same frequency for diagnostic ECG devices. Furthermore, a 300-μV impulse response has to be proven when the ST segment is investigated. In 2007 and 2009, the AHA Electrocardiography and Arrhythmia Committee, the Council on Clinical Cardiology, the American College of Cardiology Foundation, and the Heart Rhythm Society (endorsed by the International Society for Computerized Electrocardiology) published several recommendations for the standardization and interpretation of the electrocardiogram.7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 In 1990, the ad hoc writing group of the Committee on Electrocardiography and Cardiac Electrophysiology of the Council on Clinical Cardiology and the AHA published a recommendation21 stating that the lowest frequency component in the ECG corresponds to the longest RR interval. Thirty beats per minute correspond to a frequency of 0.5 Hz, 40 beats per minute correspond to 0.67 Hz. The recommendation also states that a filter with such a cutoff frequency leads to considerable phase distortions of the ECG signal, influencing the ST segment. The recommendations are based on studies done by Berson and Pipberger.22, 23 Furthermore, the possibility of digital filtering allowing zero-phase distortion in the off-line case is cited among others. Their conclusion is to use a step impulse response test to illustrate the filter's performance and guarantee optimal safety. The step impulse requirement was investigated by Berson and Pipberger.22, 23

In 1967 and 1975, the AHA recommended corresponding test methods and filter behaviors for the ECG amplifier,24, 25 as in 1990.21 In later articles, no ECG amplifier response requirements are defined; all the articles focus on lead definitions, notation, and so on. The studies of Berson and Pipberger in 1966/196722, 23 were the most in-depth studies we found. In these studies, analogue DC-coupled ECG amplifiers and analogue AC-coupled ECG amplifiers were compared (by introducing an HPF with different cutoff frequencies). Furthermore, they cite other studies on low-frequency response impacts and ST-segment distortions.26, 27, 28 Finally, in 1954, the AHA recommended a time constant of 1.6 seconds corresponding to a 0.1-Hz cutoff frequency.29

Back to Article Outline

Materials and methods 

Based on the history research, we conclude that the 0.05-Hz HPF and the impulse response test have to be investigated in more detail. Furthermore, the link to the –3- and –0.9-dB requirement has to be established. To qualitatively describe the filter behavior, we start by digitally passing through a triangular pulse (1.5-mV height) simulating a unipolar QRS complex. The first filter tested has linear phase and a high cutoff frequency (beyond the 0.05-Hz limit); the second filter has linear phase with a low cutoff frequency. In the next step, we passed the triangle through a forward-backward filter having zero phase and with a high cutoff frequency.

To quantitatively describe the filter behavior, we used the 300-μV rectangular pulse proposed by the standards. This pulse simulates the worst case scenario of a unipolar high-amplitude QRS complex.1 We passed the pulse through a digital first-order HPF with a cutoff frequency of 0.05 Hz. The same pulse was passed through an all-pass filter with phase response corresponding to the 0.05 HPF. Furthermore, the pulse was passed through a forward-backward (zero-phase) filter with different cutoff frequencies. The test pulses and the digital filter have been designed in Matlab (Mathworks, Natick, MA).

Back to Article Outline

Results 

The result of the 1.5-mV triangular pulse passed through a linear phase filter with a high cutoff frequency is shown in Fig. 1. It can be observed that a short ST-segment depression of 600 μV results. When the same pulse is passed through a filter with a low cutoff frequency, the ST depression becomes smaller but longer (Fig. 2). When the triangle is passed through a forward-backward filter, an ST-segment and PQ-segment depression results (Fig. 3).

  • View full-size image.
  • Fig. 1. 

    Forward-filtered triangle simulating the QRS complex of an ECG. The digital filter with linear phase and cutoff frequency beyond the required standards is producing an ST-segment depression. The difference of the input and output signal shown is equivalent to the suppressed signal part.

  • View full-size image.
  • Fig. 2. 

    Forward-filtered triangle simulating the QRS complex of an ECG. The digital filter with linear phase and a low cutoff frequency is producing a smaller but longer ST-segment depression that may even influence the end of the T wave. The difference of the input and output signal shown is equivalent to the suppressed signal part.

  • View full-size image.
  • Fig. 3. 

    Forward-backward, zero-phase filtered triangle simulating the QRS complex of an ECG with the same cutoff frequency. Such filter results in an ST-segment change as well as a PQ-segment change. The result of the forward filter only including the ST-segment depression is shown as well.

When the rectangular 3-μV pulse is passed through a first-order digital filter with a cutoff frequency of 0.05 Hz, the offset is 95 μV (the limit set by the standards is 100 μV). As the next step, the rectangular test pulse though could be passed through an all-pass filter corresponding to a 0.05-Hz filter. In that case, the filter does not affect the frequency content of the test pulse but the phase of the signal. We detected in this case an offset of 94 μV. When using a forward-backward filter (with zero phase), the maximal filter cutoff frequency is 0.14 Hz. In this case, the offset after a 3-μV rectangular pulse is exactly 100 μV. If the cutoff frequency of the HPF is 0.67 Hz, the offset was found to be 195 μV. This value is beyond the limit of 100 μV set by the standards.

Back to Article Outline

Discussion 

Standards, studies, and guidelines 

It is interesting that the requirements stated in the standards1, 2, 3, 4, 5, 6 are all based on the studies done by Berson and Pipberger22, 23 in 1966/1967. Furthermore, it easily can be observed that all studies between 1937 and 196726, 27, 28 use a DC-coupled analogue ECG amplifier and not an AC-coupled ECG amplifier. They tried to introduce an analogue HPF to get an AC-coupled analogue ECG amplifier, probably to avoid the manual adjustment of the ECG on the oscilloscope's knob. Moreover, it is interesting to see that Berson and Pipberger22, 23 suggest in their conclusions that the AHA recommendation (cutoff frequency of 0.1 Hz) from 195429 leads to ST-segment distortions. They further suggest that for high fidelity, DC-coupled ECG amplifiers should be used, indirectly saying that an AC-coupled ECG amplifier is not ideal. Therefore, the 0.05-Hz HPF requirement and the impulse response test were probably found to be the optimal compromise between using an HPF (doing AC coupling) and an acceptable limit of ST-segment distortions. With the introduction of ADCs (digital ECG amplifiers) in the 1980s, the AC coupling was required to reduce the large ECG offset to be able to convert the signal to the digital domain with a reasonable resolution (eg, 12 bits, 5 μV/bit gives rise to ±10.24 mV signal range before the ADC). As the number of convertible bits has increased, the possible range of conversion reaches the ECG offset limit of ±300 mV (a 20-bit ADC fulfils this requirement if a 1-μV/bit resolution is desired). And finally, their study elegantly shows that essentially monophasic QRS patterns are more likely to be recorded with distortions, indirectly saying that the worst case is not the normal ECG, but specific abnormal ECG signals such as present for example in patients with hypertrophy or found in V1/V2 or V5/V6.

Qualitative results 

The qualitative results of the 1.5-mV triangular pulse test show that a unipolar QRS complex provokes ST-segment depressions if the pulse is positive (Fig. 1). It could easily be shown that a negative pulse would provoke ST-segment elevations. As an HPF is rejecting the DC and low-frequency component of a signal, the filter rejects the mean value of the past ECG samples. This “sliding mean value” (moving average), which is the difference between the input and the output signal, is shown in Fig. 1, Fig. 2, respectively. Thus, if the complex is positive, the subtracted sliding mean is positive as well and responsible for the ST-segment depression. If the QRS complex is of negative value, the subtracted moving average value would be negative and responsible for an ST-segment elevation. The qualitative results of the 1.5-mV triangular pulse test explain the genesis of the ST-segment distortion introduced by the HPF with linear phase. By mathematics, similar results would be obtained with a non–linear-phase filter. Furthermore, our digitally found results match the result of Berson and Pipberger using analogue DC-coupled ECG amplifiers.

The zero-phase (forward-backward) filter might be a better solution, as the filter does not lead to phase distortions; only the frequency content of the signal is changed. We have shown (Fig. 3) that a forward-backward filter can lead to ST- and PQ-segment distortions. The ST-segment distortions are due to the forward filtering (identical to the above case); the PQ-segment distortions are due to the backward filtering. If the signal in Fig. 1 is time reversed, only the result of the backward filter remains. Thus, the backward filter has the same effects (flipped in time because of time reversal) as a forward filter. Instead of affecting the ST segment, the backward filter part affects the PQ segment, as the filter starts on the right and ends on the left (Fig. 3). A forward-backward filter therefore might provoke ST-segment elevation or depression amplitude measurement error due to a change of the Qonset amplitude as a result of the PQ-segment distortion of the backward part of the zero-phase filter.

Quantitative results 

We were able to show with the 3-μV impulse response test that the 0.05-Hz HPF requirement is linked to it. The impulse response tests using an all-pass filter test show the same behavior as the 0.05-Hz HPF (having in mind that an all-pass filter is only affecting the phase content and not the frequency content of the signal). Moreover, from mathematics, it is clear that the lower the cutoff frequency of such a first-order HPF is, the smaller the distortions will be. Berson and Pipberger have shown this already with analogue DC-coupled ECG amplifiers. We conclude that the frequency response test allowing –3 or –0.9 dB is not as restrictive as the impulse response test. In fact, the impulse response test is equivalent to a first-order 0.05 Hz HPF. As the frequency and the impulse response requirements need to be fulfilled at the same time, the analogue HPF is limited to its equivalence of a 0.05-Hz HPF even when the frequency requirement allows a cutoff frequency of 0.67 Hz.

Digital DC-coupled ECG amplifier 

The digital ECG amplifier, as shown, requires prior high-pass filtering, reducing the static component of the ECG signal (±300 mV) before the analogue to digital conversion (converting the dynamic part of the ECG ±10 mV) if the number of conversion bit is not high enough. In this case, it should be asked if it is possible to digitally HPF the ECG after its conversion from analogue to digital. The first right answer to this question is that, logically, it should not be possible. The reason is that, if the analogue HPF already goes to the limits of the requirement, how can it be possible to filter the ECG signal any further. To make it more complicated, in some cases, it is possible. One can try to find a perfect match between an analogue HPF with a lower cutoff frequency than 0.05 Hz and a digital filter both having an overall response fulfilling the requirements. Or one can try to use a phase equalizer that does not affect the frequency response, but compensates the phase distortions of the analogue 0.05-Hz HPF and then HPF digitally, introducing signal delay and creating the need for larger calculation resources, which are already limited. Or one can use zero-phase filtering by time reversing the signal filtering the signal with a corresponding filter (backward filtering); but this requires an almost perfect match of the digital backward filter and the analogue HPF, and we have seen that such filters affect the PQ segment as well. Knowing that analogue components never are as accurate as the coefficients in a digital filter, the 2 filters for each produced ECG device need to be matched; or other measures must be taken.

The conclusion of the above thoughts is that it would be desirable to develop a DC-coupled digital ECG amplifier, thus not having to depend on any analogue HPF (Fig. 4). We conclude that if we can merge the analogue DC-coupled ECG amplifier with today's digital technology, we can meet the problem of the 0.05-Hz HPF. Furthermore, the same methodology used before 196722, 23, 26, 28 could be used to get more fundamental knowledge about the introduction of the digital HPF in the ECG signal chain (either analogue/past days or digital/today). In such case, we would require 20 bits for the ADC, as already shown. Unfortunately, the problem of centering the ECG signal in the middle of the screen is not solved with a DC-coupled digital ECG amplifier, as there is no HPF removing the static DC signal part of the ECG. The ECG signal should somehow be centered even if the signal contains the passed ECG offset. Therefore, an HPF is still required. Nevertheless, the big advantage of the DC-coupled ECG amplifier is that this filter can be digital. This of course offers freedom. This means that the time constant can be adjusted and optimized during the complete ECG signal, allowing a large time constant during QRS complexes and other energy-rich signal parts and a smaller time constant during other parts.

  • View full-size image.
  • Fig. 4. 

    Comparison between a digital AC-coupled and a digital DC-coupled ECG amplifier. For the digital DC-coupled amplifier, the analogue HPF is replaced by a digital HPF located after the ADC. In this case, the overall signal behavior is almost identical. Both concepts, when fulfilling the corresponding standards, are safe; but the digital DC-coupled ECG amplifier gives a higher degree of freedom for future design improvements. Furthermore, the digital DC-coupled ECG amplifier allows easy testing because no HPF is present in the raw digital signal.

Furthermore, its overall behavior could be nonlinear and still include linear phase to avoid ST-segment distortions due to phase shift.22, 23 Introducing such strategies might improve the ECG signal quality, but we first have to establish their limits. And this is only possible with a DC-coupled ECG amplifier as we learned from the studies of Berson and Pipberger. The digital DC-coupled ECG amplifier can have (almost) identical performance and safety levels as the AC-coupled ECG amplifier by simulating in digital the 0.05-Hz HPF. The big advantage of the digital DC-coupled ECG amplifier is that the raw data in digital form do include the 0.05 HPF, allowing new strategies to improve the ECG device's overall performance, while safety and minimal performance requirements as defined by the current standards can be fully achieved.

Back to Article Outline

Conclusion 

Existing ECG devices that fulfil the standards are safe. The standards require a 300-μV impulse to test the ECG amplifier's HPF safe behavior. A digital AC-coupled ECG amplifier including an analogue 0.05-Hz HPF performs as well as a digital DC-coupled ECG amplifier including a digital 0.05-Hz HPF. The advantage of a digital DC-coupled ECG amplifier is that digital filter design techniques give a higher degree of freedom compared with a hardware 0.05-Hz HPF design. For example, linear phase can be obtained while retaining the cutoff frequency, which is certainly an advantage in case of the ECG (a time-coded signal). Digital filter matching is easier and safer than matching a digital filter to an analogue filter where component values naturally vary and may depend on the local print temperature.

The disadvantage of the high price for a 20- or 24-bit ADC no longer exists because the home cinema industry (eg, high-definition TV or Dolby TrueHD) uses such 24-bit ADC in audio consumer products. The increased number of produced ADC chips results in a price being in the range of a 16-bit ADC some years ago. However, the main advantage of a DC-coupled ECG amplifier is that it makes studies possible similar to the ones done before 1967, investigating the influence of the HPF on the ECG. Digital AC-coupled ECG amplifiers are inadequate for such studies. Actual standards and guidelines are based on the old studies even if, in the meantime, the development of the ECG amplifier has not stopped.

A DC-coupled digital ECG amplifier allows scientific and easy investigations of the influence of digital HPF by just comparing the digital raw signal with the high-pass filtered version of the same signal. This gives a better idea of the filter's influence on the ST segment as well as on other parts of the ECG signal. Such test can include the artificially generated signal as defined by the standards, but also real physical ECG signals. Such DC-coupled digital ECG amplifier helps improve the ECG signal quality because the methodology is quite similar to the one used when the 0.05-Hz analogue HPF requirement was investigated in the 1960s using analogue DC-coupled ECG amplifiers.

Back to Article Outline

References 

  1. IEC 60601-2-51: 2003. Medical electrical equipment—part 2-51: particular requirements for safety, including essential performance, of recording and analysing single channel and multichannel electrocardiographs.
  2. IEC 60601-2-27: 2005. Medical electrical equipment—part 2-27: particular requirements for safety, including essential performance, of electrocardiographic monitoring equipment.
  3. IEC 60601-2-47: 2001. Medical electrical equipment—part 2-47: particular requirements for the safety, including essential performance, of ambulatory electrocardiographic systems.
  4. ANSI/AAMI EC11: 1991/(R)2001. Diagnostic electrocardiographic devices.
  5. ANSI/AAMI EC-13: 1992. Cardiac monitors, heart rate meters and alarms.
  6. AAMI/ANSI EC38: 1998. Ambulatory electrocardiographs.
  7. Hancock EW, Deal BJ, Mirvis DM, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part V: electrocardiogram changes associated with cardiac chamber hypertrophy: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2009;53:992
  8. Hancock EW, Deal BJ, Mirvis DM, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part V: electrocardiogram changes associated with cardiac chamber hypertrophy: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation. 2009;119:e251
  9. Kligfield P, Gettes LS, Bailey JJ, et al. Recommendations for the standardization and interpretation of the electrocardiogram. Part I: the electrocardiogram and its technology: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Heart Rhythm. 2007;4:394
  10. Kligfield P, Gettes LS, Bailey JJ, et al. Recommendations for the standardization and interpretation of the electrocardiogram: part I: the electrocardiogram and its technology: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2007;49:1109
  11. Kligfield P, Gettes LS, Bailey JJ, et al. Recommendations for the standardization and interpretation of the electrocardiogram: part I: the electrocardiogram and its technology: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation. 2007;115:1306
  12. Mason JW, Hancock EW, Gettes LS. Recommendations for the standardization and interpretation of the electrocardiogram. Part II: electrocardiography diagnostic statement list: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Heart Rhythm. 2007;4:413
  13. Mason JW, Hancock EW, Gettes LS, et al. Recommendations for the standardization and interpretation of the electrocardiogram: part II: electrocardiography diagnostic statement list: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation. 2007;115:1325
  14. Mason JW, Hancock EW, Gettes LS, et al. Recommendations for the standardization and interpretation of the electrocardiogram: part II: electrocardiography diagnostic statement list: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2007;49:1128
  15. Rautaharju PM, Surawicz B, Gettes LS, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part IV: the ST segment, T and U waves, and the QT interval: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2009;53:982
  16. Rautaharju PM, Surawicz B, Gettes LS, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part IV: the ST segment, T and U waves, and the QT interval: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation. 2009;119:e241
  17. Surawicz B, Childers R, Deal BJ, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part III: intraventricular conduction disturbances: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2009;53:976
  18. Surawicz B, Childers R, Deal BJ, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part III: intraventricular conduction disturbances: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation. 2009;119:e235
  19. Wagner GS, Macfarlane P, Wellens H, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part VI: acute ischemia/infarction: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Endorsed by the International Society for Computerized Electrocardiology. J Am Coll Cardiol. 2009;53:1003
  20. Wagner GS, Macfarlane P, Wellens H, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part VI: acute ischemia/infarction: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation. 2009;119:e262
  21. Bailey JJ, Berson AS, Garson A, et al. Recommendations for standardization and specifications in automated electrocardiography: bandwidth and digital signal processing. A report for health professionals by an ad hoc writing group of the Committee on Electrocardiography and Cardiac Electrophysiology of the Council on Clinical Cardiology, American Heart Association. Circulation. 1990;81:730
  22. Berson AS, Pipberger HV. Electrocardiographic distortions caused by inadequate high-frequency response of direct-writing electrocardiographs. Am Heart J. 1967;74:208
  23. Berson AS, Pipberger HV. The low-frequency response of electrocardiographs, a frequent source of recording errors. Am Heart J. 1966;71:779
  24. Pipberger HV, Arzbaecher RC, Berson AS, et al. Recommendations for standardization of leads and of specifications for instruments in electrocardiography and vectorcardiography. Report of the Committee on Electrocardiography, American Heart Association. Circulation. 1975;52:
  25. Kossmann CE, Brody DA, Burch GE, et al. Recommendations for standardization of leads and of specifications for instruments in electrocardiography and vectorcardiography. Report of the Committee on Electrocardiography, American Heart Association. Circulation. 1967;35:
  26. Stoboy H, William E. Fälschungen biologischer Potentiale, insbesondere des Elektrokardiogramms, durch Verstärker und Aufzeichnungsferäte. Ztschr. Kreisluafforsch. 1954;43:
  27. Lepeschkin E. Modern electrocardiography. Baltimore: Williams & Wilkins Company; 1951;
  28. Frey A. Über die Leistungsfähigkeit verschiedener Elektrocardiographen. Ztschr. Kreisluafforsch. 1937;29:
  29. Recommendations for standardization of electrocardiographic and vectorcardiographic leads. Circulation. 1954;10:564

PII: S0022-0736(09)00324-0

doi:10.1016/j.jelectrocard.2009.07.012

Journal of Electrocardiology
Volume 42, Issue 6 , Pages 574-579, November 2009

Article Tools

* Image Usage & Resolution