MAXIMUM LIKELIHOOD ESTIMATION USING EXPECTATION MAXIMIZATION APPLIED TO AMBIENT DOSE EQUIVALENT MEASUREMENTS

Area of Science:

• Medical Physics
• Radiation Detection and Measurement
• Spectrometry

Background:

• Accurate ambient dose equivalent measurements rely on determining incident photon spectra from detector readings.
• Existing methods for unfolding pulse-height spectra often lack sufficient energy resolution.

Purpose of the Study:

• To implement and validate a maximum likelihood estimation (MLE) method for unfolding photon pulse-height spectra.
• To improve the energy resolution of incident photon spectra obtained from spectrometric detectors.

Main Methods:

• Utilized expectation maximization (EM) algorithm for maximum likelihood estimation.
• Validated the methodology using simulated and reference photon beams.
• Incorporated detector resolution into the unfolding procedure.

Main Results:

• Achieved excellent agreement between unfolded spectra and reference values within uncertainty margins.
• Demonstrated successful validation using diverse photon beam types.
• Showcased the capability of the method to account for detector resolution.

Conclusions:

• The developed MLE-based method provides a robust approach for spectral unfolding.
• This technique enhances the accuracy of ambient dose equivalent measurements.
• The improved energy resolution offers significant advantages in radiation dosimetry applications.