John A. Milliken

Electrocardiographic diagnosis of left ventricular hypertrophy in the presence of left bundle branch block: an echocardiographic study.

This study tests the electrocardiographic diagnosis of left ventricular (LV) hypertrophy in the presence of left bundle branch block (BBB). The LV mass of 125 patients with left BBB was estimated by echocardiography. M-mode echocardiography was technically adequate in 80% of patients. LV mass was calculated using previously validated M-mode formulas and then indexed to body surface area. The known shifts in the QRS voltage and axis with the onset of left BBB led to the selection of 4 electrocardiographic parameters for the diagnosis of LV hypertrophy: R in aVL 11 or more; QRS axis -40 degrees or less (or SII greater than RII); SV1 + RV5 to RV6 40 or more; SV2 30 or more and SV3 25 or more; these parameters were used in cumulative fashion. This cumulative approach was superior to using single conventional criterion such as the SV1 + RV5 or RV6. When LV hypertrophy was defined as an M-mode index of at least 115 g/m2, the sensitivity was 75% and specificity 90%. Using an M-mode mass of at least 215 g as the standard, the sensitivity was 73% and the specificity 66%. LV hypertrophy can be diagnosed by electrocardiographic criteria in the presence of left BBB at least as reliably as in normal conduction.

Isolated and complicated left anterior fascicular block: A review of suggested electrocardiographic criteria

The electrocardiographic criteria for isolated left anterior fascicular block are reviewed and illustrated. Left anterior fascicular block decreases the voltage in the chest leads and increases the voltage in the limb leads. The usual voltage criteria of left ventricular hypertrophy must be modified appropriately. Changes in repolarization include a decrease in the T wave of leads I and AVL and an increase in leads II, III, AVF, V5 and V6. Small Q waves in V2 may simulate an anteroseptal myocardial infarction. Three criteria for the possible diagnosis of superimposed inferior myocardial infarction have been suggested. Both inferior and anterior myocardial infarctions may be masked by R waves replacing Q waves. In the presence of a recent anterior infarction, right bundle branch block may also be masked. Thus, left anterior fascicular block may mask or mimic infarction and left ventricular hypertrophy and mask right bundle branch block in the setting of an acute anterior myocardial infarction.

Prediction of left ventricular mass from the electrocardiogram

Multiple stepwise regression methods were used to derive electrocardiographic (ECG) models for prediction of the echocardiographic left ventricular (LV) mass index from standard 12-lead ECG measurements using data files of 203 men and 252 women. The correlation between echocardiographic and ECG estimates of LV mass index was R2 = 0.58 for men and R2 = 0.42 for women. A separate logistic regression model was derived for classification of LV hypertrophy as a dichotomized dependent variable. This classifier chose R (aVL), T (V6), and S (V1) for men and R (aVL), T (V6), and S (I) for women and produced a moderate sensitivity (53.7% for men and 63.4% for women) and specificity (94.9% for men and 92.9% for women). We conclude that the initial performance of these and other recently developed multivariate estimators of LV mass and LV hypertrophy classifiers is promising enough to subject them to further studies to evaluate their utility as risk predictors.

Computer program for pattern recognition of electrocardiograms

Abstract A computer program for the ECG pattern recognition by a computer has been described. It utilizes a thresholding principle for locating the QRS complex and employs critical points for determining its configuration. Those points indicate either change of direction or maximal slope of the ECG signal. Threshold selection and finding critical points is accomplished by calculating first differences between consecutive ECG data points. Configuration of the QRS complex is determined by fitting straight lines to critical points according to rules set in a decision table. After the baseline has been established by the “histogram” method or by a straight line fitting between points immediately before QRS complexes in two consecutive cycles, the P wave and T wave are detected on the basis of maximum or minimum amplitude and zero crossings (referred to zero DC level). Configuration of the P or T wave is established on the basis of a number of peaks, their polarity and a sequence of their occurrence.

Computer program for diagnostic evaluation of electrocardiograms

Abstract A computer program which utilizes decision tables to arrive at ECG diagnosis has been described. Given an array containing the amplitudes and durations of the waveform components of the 12-lead system ECG the program matches them against standard cardiological criteria that have been set in a series of decision tables linked together by a “GO TO …” statement. The first decision made is between arrhythmia and regular rhythms. Then a series of decision tables is processed and a tentative diagnosis, to be checked by a cardiologist, is printed.

Logical approach to diagnosing electrocardiograms

Summary A method is proposed in which the diagnostic process in electrocardiography is represented by a series of questions, answers and diagnostic statements arrayed in a tree-like structure or a decision table. In the first case every question-node has two answer-branches, YES and NO, which lead to other questions. The final answers are terminated by diagnostic statements. In the decision table, questions and diagnostic statements are listed in the left hand section, while the answers to those questions and the diagnosis to be made appear in the right hand section. This method not only aids the physician to evaluate the ECG data in an efficient and prompt way, but also may be utilized for a computer-assisted intepretation of electrocardiograms.

The impact of the computer-reported vectorcardiogram on the cardiologist interpreter of the scaler electrocardiogram*

The purpose of this study was to determine if the computer-reported vectorcardiogram (VCG) had a notable impact on the cardiologist interpreter of the 12-lead scalar electrocardiogram (ECG). Three cardiologists read 100 12-lead scalar ECGs and four months later again read the same tracings while having available the VCG computer report. The diagnosis was altered in 25% of the repeated interpretations. Forty-five per cent of these had only minor changes and can probably be disregarded. Major or significant changes occurred in 14% of the records, and 80% of these were apparently attributable to the computer report. It was concluded that the use of a computer-assisted interpretation of a VCG may enhance the uniformity and consistency of the cardiologists interpretation of the scalar ECG.