Long-term stability evaluation of user-derived electrometer calibration factors (kelec, user) for radiotherapy dosimetry

Abstract

Purpose

User-derived electrometer calibration factors (k_{elec, user}) have been proposed as a practical approach for electrometer quality assurance in radiotherapy dosimetry. However, the contribution of long-term stability to the uncertainty of k_{elec, user} has not been quantitatively evaluated. This study aimed to assess the long-term stability of k_{elec, user} over a 28-month period and to validate the method using different signal input techniques.

Methods

A RAMTEC Solo electrometer was used as the reference electrometer, and three electrometers (UNIDOS^webline, RAMTEC Duo, and Super MAX) were evaluated as electrometers under test. Electrometer sensitivity ratios were measured using three signal input methods: A LINAC-based method, a battery-powered current source (SCG002), and a servo-controlled current source (EMF521R). User-derived calibration factors were calculated from the sensitivity ratios and the calibration coefficient of the reference electrometer. Long-term stability was evaluated from repeated measurements over 28 months using linear regression analysis, and the associated uncertainty component was estimated according to the Guide to the Expression of Uncertainty in Measurement (GUM).

Results

The coefficients of variation for repeated measurements were ≤ 0.071% for all methods. The k_{elec, user} values obtained using the LINAC method, SCG002, and EMF521R agreed within 0.1%, with a maximum relative difference of 0.06%. Over the 28-month observation period, the largest estimated drift was 1.84 × 10⁻⁴, corresponding to a Type B standard uncertainty of 1.06 × 10⁻⁴ (1.06 × 10⁻²%). The combined standard uncertainty of k_{elec, user} remained 0.20%, and the experimentally determined long-term stability component had a negligible effect on the overall uncertainty.

Conclusions

The influence of signal input method on k_{elec, user} was minimal, and consistent results were obtained using LINAC-based and current-source-based measurements. The uncertainty contribution from long-term stability over 28 months was negligible compared with the reported uncertainty of electrometer calibration factors. These findings support the use of periodic in-house sensitivity comparison measurements as a complementary quality assurance tool for monitoring electrometer performance between external calibrations. These findings extend previous studies by providing quantitative evidence of the relative sensitivity stability of user-derived electrometer calibration factors and thereby support their application for routine in-house quality assurance.