Reply to the comment on ‘Monte Carlo calculated microdosimetric spread for cell nucleus-sized targets exposed to brachytherapy ¹²⁵I and ¹⁹²Ir sources and ⁶⁰Co cell irradiation’

Abstract

A discrepancy between the Monte Carlo derived relative standard deviation (microdosimetric spread) and experimental data was reported by Villegas et al (2013 Phys. Med. Biol. 58 6149–62) suggesting wall effects as a plausible explanation. The comment by Lindborg et al (2015 Phys. Med. Biol. 60 8621–4) concludes that this is not a likely explanation. A thorough investigation of the Monte Carlo (MC) transport code used for track simulation revealed a critical bug. The corrected MC version yielded values that are now within experimental uncertainty. Other microdosimetric findings are hereby communicated.

Microdosimetric considerations for radiation response studies using Raman spectroscopy

“…We also investigate the relationship between the microdosimetric spread (the relative standard deviation of the specific energy; ), dose D in Gy, and target volume V in μm 3 using the following equation:which is based on equations (2.19) and (A.12) from ICRU Report 36, and is in agreement with Villegas et al. The curve_fit function from the optimize package of the SciPy library of the Python programming language (http://www.python.org) is used to determine different C 1 and fit parameters for each incident photon spectrum.…”

“…Our simulation approach involved using EGSnrc, a class II condensed history MC code, to score specific energy within populations of μm‐scale targets. Villegas et al used a different approach to estimate the microdosimetric spread. They used a modified version of PENELOPE to carry out track structure (TS) MC simulations to calculate the single track frequency distribution, f 1 ( z ), which was repeatedly convolved with itself to obtain the (multi‐event) specific energy distribution at the desired dose level.…”

“…They used a modified version of PENELOPE to carry out track structure (TS) MC simulations to calculate the single track frequency distribution, f 1 ( z ), which was repeatedly convolved with itself to obtain the (multi‐event) specific energy distribution at the desired dose level. Their initial work considered cubic voxel targets, but these results must be disregarded due to a critical bug in their MC transport code . They later considered spherical targets, providing fit parameters to Eq.…”

“…Their initial work considered cubic voxel targets, but these results must be disregarded due to a critical bug in their MC transport code. 44 They later considered spherical targets, providing fit parameters to Eq. ( 2) for 125 I, 192 Ir and 1.25 MeV photons (with the latter approximating a 60 Co spectrum).…”

Deep learning-based tumor segmentation on digital images of histopathology slides for microdosimetry applications

“…Currently, in microdosimetry calculations performed with the Monte Carlo (MC) method, properties such as energy absorption have been studied using a target geometry made up of single-sized volume mimicking individual cells. [10][11][12][13][14][15][16] To represent nucleus sized targets, some authors use smaller volumes, which reside within the cell body. 11,13 While using uniformly sized cells and nuclei as a target geometry may provide useful information in many dosimetry applications, any effects that a nuclei population with diverse volume size distributions may have on the studied quantities are usually ignored due to lack of available data.…”