Search Results

You are looking at 1 - 4 of 4 items for

  • Author: D. Rodbard x
  • All content x
Clear All Modify Search
Restricted access

A. D. Genazzani and D. Rodbard


We utilize the "Receiver Operating Characteristic" to describe the relationship between sensitivity and specificity as the threshold for peak detection is varied systematically, to provide objective comparison of the performance of methods for detection of episodic hormonal secretion. A computer program was used to generate synthetic data with peaks with variable durations, with constant or variable height, shape and/or interpulse interval. This approach was used to compare the CLUSTER and DETECT programs. For both programs, the observed false positive rates estimated using signal-free data were in good agreement with the nominal rates, but in the presence of signal the observed false positive rates were systematically lower. Sensitivity increases with increasing signal/noise ratio, as expected. Program DETECT, using its standard options, provided excellent sensitivity (90-100%) with very low false positive rate under all conditions tested. Its performance could be further improved by the use of a more stringent definition of a peak requiring the presence of "UP" followed by a "DOWN". The CLUSTER program was found to have very poor sensitivity when using the "local variance" option. Use of the true fixed standard deviation or percent coefficient of variation resulted in a modest improvement. Optimal performance of program CLUSTER was obtained by the use of the best of 3 variance models, testing 12 different cluster sizes (from 1×1) to 4×4 and selecting the best among these: under these conditions it can achieve high sensitivity (90-100%) for very low observed false positive rate, such that its performance was comparable to that of DETECT. The methods developed and illustrated here should permit the definitive characterization and validation of the performance of any one method, the objective comparison of the relative performance of two or more methods for analysis of pulsatile hormone levels for episodic hormone secretion, and lead to the improvement of algorithms for peak detection.

Restricted access

D. Rodbard and J. E. Lewald


A computer program written in extended BASIC (BIICAC) is described for analysis of radioimmunoassay and competitive protein binding assay data. This method uses logit and log transformations to obtain a linear dose responsive curve, followed by unweighted and iterated weighted least squares regression analysis (similar to probit analysis). The standard curve is plotted by computer, and potency estimates and confidence limits are obtained for unknowns. This program has been used successfully in a fully automated data processing system using a time-sharing computer service.1

Restricted access

Vincenzo Guardabasso, Alessandro D. Genazzani, Johannes D. Veldhuis, and David Rodbard


A new objective method is presented for investigating the presence of a temporal relationship between episodic release of two hormones. The two time series of hormone concentrations are first analysed by an objective method for peak detection. Both data series are then transformed into "quantized" or discretized series by recording the occurrence of a hormone pulse as an "event", characterized by the onset, the maximum, or another unique feature. The two quantized series are then matched, and the number of concordant events and discordant events are counted. Each point in series A is compared with a "time-window" of a selected number of points in series B, to accommodate small degree of mismatch between events in the two series. An index of concordance is computed, compensating for any spurious random coincidence: the "Specific Concordance", to evaluate the frequency of concordant events in excess of those expected on the basis of chance alone. This calculation is systematically repeated, interposing a range of time-lags between the two series. A graph of Specific Concordance versus time-lag indicates the time-lag corresponding to a maximal concordance. Simulations of random series of events are performed, and their degree of concordance is evaluated in a similar fashion, thus generating frequency distributions of Specific Concordance values under the null hypothesis of no temporal relationship. This permits the selection of criteria for statistical significance at any desired p-level, for one or many lag times, and for one or multiple subjects. Various degrees of concordance can also be simulated to evaluate the performance (sensitivity, statistical power) of this approach. These methods have been implemented as a collection of short microcomputer programmes, and applied to the study of the temporal relationship between β-endorphin and cortisol in normal subjects sampled every 10 min for 24 h. This analysis demonstrated concordance between events in the two series, with synchronous occurrence of β-endorphin and cortisol release events significantly more frequently than expected on the basis of random association (p<0.01).