Welch Allyn Cardioperfect Workstation, PEDMEANS ECG Interpretation Module - Physicians - User Manual - Manual (Page 7)

CardioPerfect Workstation
PEDMEANS ECG Interpretation Module Physicians Manual
DIR 80015051 Rev. C 7 / 55
1.4 Diagnostic classification
The diagnostic logic operates on the parameters and produces both a rhythm classification and a
contour or morphology classification. The criteria used by the computer may differ from the criteria
used in the ECG textbooks. The basic reason is that a human observer is inaccurate but flexible
and creative, a computer precise and obedient but rigid in its operation.
There are several specific reasons why ECG criteria in the program may differ from the
conventional ones. First there is no uniformity of criteria in the literature. Then, criteria may be
based on inaccurate measurement by eye. Also, ECG measurements may be “falsified” for the
ease of the reader: axis calculations are generally made from the amplitudes of QRS complexes
rather than from the surface areas under the QRS tracings as prescribed by theory. Further, criteria
are sometimes not quantitatively defined (How flat must a flat ST-T be? How slurred is a slurred
QRS upstroke?) Or their measurement is not unequivocally prescribed. For the computer program
to work, a quantitative definition must somehow be decided upon. Moreover, conventional criteria
may have been based on measurements produced by technically outdated instrumentation. The
amplitudes of R waves have been consistently underestimated, especially in children, due to
filtering effects by too low frequency response of the electrocardiographs. Finally, a human
interpreter may deviate from strict criteria as he sees fit: sometimes criteria have been made to
meet a priori expectations.
In one respect the computer is inferior to the human observer: although the computer can measure
very accurately, its powers of pattern recognition are inferior. For instance, it will have great trouble
in detecting a P wave buried in a ST segment which is easily seen by the human eye.
Diagnostic interpretation of pediatric ECGs, much more than adult ECGs, relies on the use of age-
dependent normal values in the classification rules. Normal limits of the pediatric ECG have been
established in the past, but each of these studies has its deficiencies. Therefore, during the
development of the interpretation program described in this manual a new set of normal limits was
established using a data set of about 2,000 ECGs from normal children aged 0 to 16 years.
Continuous age-dependent curves were calculated for the upper limit of normal (taken as the 98th
percentile) and the lower limit of normal (2nd percentile), for all parameters used in the diagnostic
interpretation. These curves avoid abrupt changes in diagnosis with small differences in age. In
Appendix A approximations of the continuous age-dependent normal limits are presented in tabular
form.
1.5 Outline of the manual
This manual consists of two main parts. One part describes the diagnostic criteria that are
employed in the contour classification of the Pediatric Modular ECG Analysis System
(PEDMEANS), the other describes the criteria used in the rhythm classification of PEDMEANS.
Each part contains a brief introductory section, a description of the measurements that are used in
the diagnostic logic, and a comprehensive list of statements and corresponding diagnostic criteria.
Related statements have been grouped in sections, e.g., all statements related to intraventricular
conduction delay, left ventricular hypertrophy, etc. Finally, an index of the statements that can be
generated by the program is provided.
For each statement that can be issued, the criteria are specified in a general format. For example:
Say: “WPW”
if: delta wave in at least 2 leads
and PR interval > ULN