A strategy for finding the optimal deconvolution estimates for hormone secretory kinetics using AutoDecon

Area of Science:

• Endocrinology
• Pharmacokinetics
• Biostatistics

Background:

• Accurate estimation of hormone secretion kinetics is crucial for understanding physiological processes.
• Current deconvolution methods may lack precision, particularly with short hormone half-lives.

Purpose of the Study:

• To introduce AutoDecon, a novel automated deconvolution procedure for hormone secretion kinetics.
• To enhance the accuracy of hormone secretion kinetic estimation using an optimized model selection strategy.

Main Methods:

• Developed a multiparameter deconvolution procedure (AutoDecon).
• Implemented a corrected Akaike's information criterion for optimal model selection.
• Utilized simulated cortisol time series for validation.

Main Results:

• AutoDecon demonstrated improved accuracy in estimating hormone secretion kinetics compared to standard methods.
• The proposed method showed particular benefits in scenarios with diminished hormone half-life.
• Proof of principle for the enhanced deconvolution strategy was established.

Conclusions:

• AutoDecon offers a robust, automated approach for analyzing hormone secretion dynamics.
• The model selection strategy improves deconvolution accuracy, especially for short half-life hormones.
• Further validation and elaboration are recommended for broader clinical application.